{
  "schema": "proofloop-prod-proxy-benchmark-matrix-v1",
  "generatedAt": "2026-07-05T20:36:56.832Z",
  "baseUrl": "https://noderoom.live",
  "models": [
    "z-ai/glm-5.2",
    "deepseek/deepseek-v4-flash",
    "poolside/laguna-xs-2.1",
    "qwen/qwen3.7-plus"
  ],
  "summary": {
    "uniqueTaskTargets": 1354,
    "modelCount": 4,
    "matrixAttemptTargets": 5416,
    "prodLiveBrowserVerifiedTaskTargets": 4,
    "localLiveBrowserOnlyTaskTargets": 101,
    "runnableProdBrowserTaskTargets": 1354,
    "blockedTaskTargets": 0,
    "prodLiveBrowserAttemptPasses": 10,
    "allTasksProdVerified": false
  },
  "recommendation": {
    "allTaskWinner": null,
    "currentProdAdapterSmokeWinner": "poolside/laguna-xs-2.1",
    "basis": "No all-task model winner is claimed until every tracked task target is run through the prod browser matrix. Current winner is limited to the completed external-adapter prod smoke."
  },
  "modelSummaries": [
    {
      "modelId": "z-ai/glm-5.2",
      "prodAdapterSmokePassed": 3,
      "prodAdapterSmokeTotal": 3,
      "estimatedCostUsdAtOpenRouterList": 0.245418,
      "measuredCostUsd": 0.245,
      "avgDurationMs": 128530.67
    },
    {
      "modelId": "deepseek/deepseek-v4-flash",
      "prodAdapterSmokePassed": 1,
      "prodAdapterSmokeTotal": 3,
      "estimatedCostUsdAtOpenRouterList": 0.068253,
      "measuredCostUsd": 0.069,
      "avgDurationMs": 475826.67
    },
    {
      "modelId": "poolside/laguna-xs-2.1",
      "prodAdapterSmokePassed": 3,
      "prodAdapterSmokeTotal": 3,
      "estimatedCostUsdAtOpenRouterList": 0.032628,
      "measuredCostUsd": 0.033,
      "avgDurationMs": 107164.33
    },
    {
      "modelId": "qwen/qwen3.7-plus",
      "prodAdapterSmokePassed": 3,
      "prodAdapterSmokeTotal": 3,
      "estimatedCostUsdAtOpenRouterList": 0.199433,
      "measuredCostUsd": 0.199,
      "avgDurationMs": 191148.67
    }
  ],
  "families": [
    {
      "id": "spreadsheetbench-v1-full-912",
      "title": "SpreadsheetBench V1 full benchmark",
      "taskCount": 912,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 1,
      "runnableProdBrowserTasks": 912,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "102-20",
          "title": "I need guidance with reformatting my Excel raw data using VBA. Specifically, I want to combine and sum up data from a co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "102-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/102-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/102-20/evaluator/evaluator.json",
            "docs/eval/spreadsheetbench-live-room-proof.json"
          ],
          "blockers": [
            "Existing SpreadsheetBench receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "10281",
          "title": "I want to configure column 'J' under 'Densities' in my Excel sheet to return 'PASS' or 'FAIL' similar to what I have set",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "10281",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10281/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10281/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "10452",
          "title": "I need a method to perform a vertical lookup in Excel that only returns values which begin with 'PK', eliminating the ne",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "10452",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10452/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10452/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "10493",
          "title": "I am working on a salary spreadsheet for my wife's new job where she needs to calculate the different denominations of c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "10493",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10493/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10493/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "105-13",
          "title": "How can I use VBA to filter data from a multi-column range on a 'Source' worksheet based on criteria in a specific cell ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "105-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/105-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/105-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "105-24",
          "title": "In my Excel file, the 'Open' sheet contains IDs in Columns C and K, and I need Excel to find all these IDs throughout th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "105-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/105-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/105-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "10660",
          "title": "How can I count the number of unique names in Column C that correspond with a specific date in Column B (as shown in H1)",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "10660",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10660/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10660/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "10747",
          "title": "I've attached a file containing a basic table and another table showing only a portion of that basic data. I've also inc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "10747",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10747/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/10747/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "108-24",
          "title": "How can I create a macro to extract rows from Sheet BZTB, where descriptions in column B start with 'COS Pass，COS Comm，C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "108-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/108-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/108-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "109-21",
          "title": "I need a macro to process a list, where groups are up to five people each. These people have point totals in column K. T",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "109-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/109-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/109-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "110-2",
          "title": "Can I use Excel VBA to apply an AutoFilter on three separate columns using three different arrays, with each array corre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "110-2",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/110-2/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/110-2/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "110-6",
          "title": "I have a workbook with two sheets named 'Distribution' and 'Calculation'. I came across a VBA code that was intended to ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "110-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/110-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/110-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "11276",
          "title": "I am using the TEXT formula in Column F4 in my Excel sheet, which is meant to return the weekday name, but it is incorre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "11276",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/11276/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/11276/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "115-31",
          "title": "How can I transfer data from the 'CS' sheet to the 'S' sheet and remove the brackets from the values using VBA code, ens",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "115-31",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/115-31/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/115-31/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "118-10",
          "title": "How can I extract data in the B column, which is right after each instance of the word 'Ledger:', to the H column, corre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "118-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "118-50",
          "title": "I need a macro to find words in an Excel file with a specified pattern where the last three letters of a word, such as '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "118-50",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-50/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-50/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "118-8",
          "title": "I need guidance on creating a database from an Invoice Template sheet in Excel, so that I can save invoice details by cl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "118-8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-8/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/118-8/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "11842",
          "title": "How can I use the COUNTIFS formula, or possibly another formula, to calculate the full-year attendance for each individu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "11842",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/11842/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/11842/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "120-24",
          "title": "I need a VBA code to validate data and replace values in Excel based on multiple conditions.  Specifically, the criteria",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "120-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/120-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/120-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "122-39",
          "title": "How can I sort duplicate names in an Excel sheet so that they are grouped together and ordered from ascending to descend",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "122-39",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/122-39/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/122-39/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "122-5",
          "title": "How can I amend my VBA macro to delete rows in Column L of a sheet named 'PL Recon Items', where there are numeric value",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "122-5",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/122-5/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/122-5/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "12307",
          "title": "How can I calculate the number of countries each company operates in (meaning they have more than 0 employees in that co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "12307",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/12307/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/12307/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "124-46",
          "title": "In my Excel workbook, after processing through our in-house software, two issues arise: firstly, in Column C, the networ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "124-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/124-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/124-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "124-9",
          "title": "How can I use VBA to replace all instances of 'Catalog Task' in Column B with 'Request', delete all rows where data in c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "124-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/124-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/124-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "127-32",
          "title": "How can I adjust my VBA code to replace or delete a specific word followed by varying numbers in an Excel dataset? This ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "127-32",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/127-32/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/127-32/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "12864",
          "title": "I am attempting to use the Index/Match function combination to retrieve corresponding dates from Column B on sheet 2 by ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "12864",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/12864/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/12864/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "13-1",
          "title": "How can I combine data from a 'RANGES' sheet to a 'LISTS' sheet, by matching duplicates based on the 'DATE' and 'REF' co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "13-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "130-3",
          "title": "In my Excel file, I need a macro that can transfer certain data from one sheet to another. Specifically, I need to ident",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "130-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/130-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/130-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "130-9",
          "title": "How can I edit and correct my incomplete Excel VBA code to combine data from two sheets where data under the same headin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "130-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/130-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/130-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "131-38",
          "title": "In my VBA code, when I try to insert a column and then sum adjacent cells across multiple rows, the formula that gets in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "131-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/131-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/131-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "132-4",
          "title": "I want to keep rows containing the name 'June' in column B of my Excel sheet and delete all the other rows, regardless o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "132-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/132-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/132-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "13284",
          "title": "How can I return a name from the 'Streets' tab, column F, into the 'Base' tab, column E, based on two criteria: one from",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "13284",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13284/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13284/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "134-14",
          "title": "How can I create a macro to delete rows in an Excel spreadsheet that contain tick marks in column B?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "134-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/134-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/134-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "13894",
          "title": "How can I create a formula in Excel that assigns a unique code number (UCODENUMBER) to each different combination of pro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "13894",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13894/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/13894/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "139-16",
          "title": "I need a macro that searches for rows in Column A containing the term 'Intagible Assets' and checks if the corresponding",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "139-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/139-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/139-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "141-20",
          "title": "How can I create a macro that finds and deletes rows in two Excel sheets named 'PL Recon Items' and 'Statement Recon Ite",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "141-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/141-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/141-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "142-12",
          "title": "I need a VB (Visual Basic) code that will iterate through Column F and Column J in an Excel spreadsheet. The code should",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "142-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "142-19",
          "title": "I have an Excel sheet titled 'sheet1' with 500 total rows, which include blank rows interspersed throughout. In this she",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "142-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "142-32",
          "title": "I need help amending my VBA code that is intended to compare and extract non-matching items between two sheets. On the '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "142-32",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-32/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/142-32/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "14207",
          "title": "How can I obtain a value in column B if any cell within the range D2:D8 matches any cell in column A in a similar way to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "14207",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14207/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14207/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "14240",
          "title": "I'm having trouble with a SUMPRODUCT formula in Excel. I've noticed an error but can't seem to figure out what's wrong. ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "14240",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14240/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14240/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "145-13",
          "title": "I need a macro code for the Excel file I have provided which displays the number of leave days taken. This should be cal",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "145-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/145-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/145-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "146-49",
          "title": "I need to identify pairs of words from a dictionary, where one word is the same as the other, except that all instances ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "146-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/146-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/146-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "14625",
          "title": "How can I correct an 'average if' function in Excel that is returning an incorrect value? Attached is a file with the sa",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "14625",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14625/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/14625/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "147-48",
          "title": "I want to create a macro that can transpose data similar to how I manually transposed data from Column A to Column C onw",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "147-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/147-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/147-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "147-7",
          "title": "I'm editing an old code I got from a user named @sintek to remove duplicates from a dataset, but I need help modifying i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "147-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/147-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/147-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15027",
          "title": "How can I shorten the existing formula that I used in cells B6:H6 in my Excel spreadsheet, which is also related to the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15027",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15027/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15027/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "151-13",
          "title": "In the VBA code of my Excel sheet, I need to modify the macro labeled 'Country'. Currently, the macro searches for occur",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "151-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/151-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/151-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1530",
          "title": "How can I create a formula in Excel that will input into Cell G1 the corresponding date value from Column A when an exac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1530",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1530/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1530/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15380",
          "title": "How can I extract a specific part of a text string from cell A3, search for it in a separate range 'codes2', find the co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15380",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15380/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15380/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15387",
          "title": "I am working on a model in Excel and need to find a formula that allows me to obtain a list of customer names that I hav",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15387",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15387/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15387/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "156-14",
          "title": "I attempted to write VBA code that would delete rows from row 6 onward if the cell in column F contains the text \"to be ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "156-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/156-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/156-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "156-8",
          "title": "I need a macro that can process my data, replicating the results in columns J to S based on the data in columns A to H a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "156-8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/156-8/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/156-8/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1563",
          "title": "In an Excel column (column A), I have a mixture of numeric and characters with some blank cells. How can I fill these bl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1563",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1563/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1563/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15671",
          "title": "I have an Excel sheet with 269 rows of data in Column A, which I have changed from number format to text format. Current",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15671",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15671/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15671/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "157-37",
          "title": "I need assistance with writing a macro that will delete the contents of cells in Columns B, C, D, and E on a row only if",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "157-37",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/157-37/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/157-37/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "157-4",
          "title": "I need to create a macro that will delete entire rows within the range of F14 to M83 on my worksheet if all the cells in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "157-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/157-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/157-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15825",
          "title": "I am seeking assistance to create a formula that will allow me to filter a list in an Excel workbook. The unfiltered dat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15825",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15825/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15825/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15827",
          "title": "How can I determine the highest price for a test code, identify which center (centre name required) it is associated wit",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15827",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15827/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15827/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "15829",
          "title": "How can I create an Excel formula to look up a tariff from a selection and cross reference it with a usage value to calc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "15829",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15829/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/15829/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "159-7",
          "title": "I am using J. Topley's macro and need a modification to achieve specific search results using patterns. When I input 'PA",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "159-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/159-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/159-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "160-6",
          "title": "I have a dataset in an Excel sheet named \"SH\" that contains blank rows interspersed among the data. I need a macro that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "160-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/160-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/160-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "164-22",
          "title": "I need a macro to compare two Excel sheets, YEAR1 and YEAR2, using column B as the basis for comparison. For each ID in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "164-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/164-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/164-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "165-23",
          "title": "I need assistance with two VBA macros for Sheet1 in Excel. The first macro, named delrow1, should delete rows where the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "165-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/165-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/165-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "16511",
          "title": "How can I create a formula in Excel to increment track numbers for songs that correspond to their respective album names",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "16511",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/16511/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/16511/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "166-27",
          "title": "How can I write a VBA macro that transposes rows of data, which are separated by blank lines, from multiple worksheets i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "166-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/166-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/166-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "168-17",
          "title": "How can I write a macro in Excel to locate the first instance of 'Invoice No.' in column A on the 'Statement' sheet and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "168-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/168-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/168-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "17-35",
          "title": "How can I successfully filtering dates in Excel based on the start and end date criteria entered in cells I2 and J2? I h",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "17-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "170-13",
          "title": "How can I check if a header from Sheet2 exists in the Entry column of Sheet1, and if it does, insert the same number of ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "170-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/170-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/170-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "17047",
          "title": "How can I find the minimum value in a specific range within an Excel spreadsheet, and what are some of the formulaic sol",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "17047",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17047/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17047/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "17049",
          "title": "I am seeking a non-array formula to perform a VLOOKUP that retrieves the most recent date related to a specified lookup ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "17049",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17049/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17049/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "17111",
          "title": "I need to sum the values in Excel based on the status. For example, for a person named 'Neha' and a type 'HD', I want to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "17111",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17111/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17111/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "172-10",
          "title": "How can I write a VBA script to filter and split data in an Excel spreadsheet where data from Column T equals \"New York\"",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "172-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/172-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/172-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "17251",
          "title": "How can I use an Excel formula to sort a list of parts in column A by their respective prices in column B so that after ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "17251",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17251/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/17251/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1726",
          "title": "How can I adjust my Excel formula to leave cells in column A6:A36 blank if their values are not within the date range sp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1726",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1726/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1726/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "175-10",
          "title": "How can I copy columns A:F in Excel and delete rows where the 'BUYING' & 'SELLING' columns do not contain either zero or",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "175-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/175-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/175-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "177-6",
          "title": "I'm seeking a macro that can merge entire rows in an Excel sheet based on matching values in a specific column. In my ca",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "177-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/177-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/177-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "178-22",
          "title": "I am attempting to use VBA to filter two columns simultaneously in Excel. In column B, I want to filter for the value 'T",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "178-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/178-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/178-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "18137",
          "title": "How can I summarize the count of unique values within a specified time range between two timestamps in Excel?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "18137",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18137/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18137/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1818",
          "title": "I have a spreadsheet tracking students' yearly performance and need a formula to identify the lowest performing students",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1818",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1818/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1818/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "183-8",
          "title": "How can I calculate the weighted average for columns J, K, and L? I have provided a sample file for reference.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "183-8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/183-8/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/183-8/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "18385",
          "title": "How can I automatically count the correct answers for each student in Excel without having to manually correct each shee",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "18385",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18385/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18385/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "184-45",
          "title": "How can I use VBA to compare data from two separate Excel sheets and output the results to a new sheet? Specifically, I ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "184-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/184-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/184-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "184-6",
          "title": "I need a macro code that counts the number of dates listed in each row from Column E to the end of each row in the 'Rawd",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "184-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/184-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/184-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1845",
          "title": "How can I adjust my Excel formula to maintain a balance sequence where entrances are treated as positive values and exis",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1845",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1845/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1845/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "185-13",
          "title": "How can I use VBA to remove duplicate names in a sheet, given that the data is formatted with each person's set of shift",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "185-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/185-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/185-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "18645",
          "title": "How can I create a calculation in Excel for weighted revenue where the percentage used to multiply the revenue varies au",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "18645",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18645/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18645/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "188-39",
          "title": "How can I update a result sheet in Excel to display the name, contact details, and email of a person when their designat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "188-39",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-39/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-39/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "188-45",
          "title": "How can I use VBA to find any instance of a word from a list of 1 to 6 words located in column H of Sheet1 within the da",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "188-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "188-46",
          "title": "How can I use VBA to update cells in columns F and G based on a set of conditions? Specifically, if column G contains 'P",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "188-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "188-47",
          "title": "I need assistance with a Visual Basic for Applications (VBA) code to manipulate data in an Excel spreadsheet. My dataset",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "188-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/188-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "189-9",
          "title": "i need a vba code to fill dates with timestamp \"=TEXT(TODAY()-1,\"dd-mm-yyyy\")&\"11:59:00 PM\"`result as 11/01/2022 11:59:0",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "189-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/189-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/189-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "18935",
          "title": "How can I perform a lookup in Excel to retrieve a value from a Data Table and place it into a report format, using the d",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "18935",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18935/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/18935/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "19-16",
          "title": "How can I modify a VBA macro so that it retains only the rows containing \"TSCO REQ\" in column H and deletes all other ro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "19-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "19-7",
          "title": "How can I create a macro to split data from the 'LASTROW' sheet into two separate sheets named 'MINUS' and 'PLUS', based",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "19-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "19032",
          "title": "How can I formulate a calculation in Excel to determine if an employee was working during a specific quarter given their",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "19032",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19032/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/19032/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "191-40",
          "title": "I need a macro that can sort varying blocks of numbers within an Excel worksheet by column D in descending order. The pr",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "191-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/191-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/191-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "192-22",
          "title": "I need a VBA code to check each cell in column D for specific keywords and to output 'Billing PO' in the corresponding c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "192-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/192-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/192-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "1925",
          "title": "I need to create a dynamic Index in Excel that changes according to the date in Sheet1, Cell C5, so that it corresponds ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "1925",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1925/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/1925/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "193-42",
          "title": "I'm seeking assistance with creating a macro in Excel. The macro should compare cells within column B and identify any r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "193-42",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/193-42/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/193-42/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "194-19",
          "title": "I am looking for a macro that can search through data on sheet two and generate results in only two columns, H and I, on",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "194-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/194-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/194-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "196-35",
          "title": "I have an Excel list of order numbers each associated with item barcodes, and I would like to select or filter multiple ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "196-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/196-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/196-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "200-27",
          "title": "I need to create a macro that will remove duplicate records from a list. These duplicates are defined by having identica",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "200-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/200-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/200-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "2015",
          "title": "How can I determine and get the index of the first column for each row in an Excel spreadsheet when the sum of the value",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "2015",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2015/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2015/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "203-15",
          "title": "I'm using Excel 2013 and have a main sheet that has data for staff, shown in the attached file 'Transpose Staff Data.xls",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "203-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "203-16",
          "title": "I need a macro that counts the words in each new row above an empty row in an Excel spreadsheet. I have manually entered",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "203-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "203-33",
          "title": "I'm looking for a code that can merge cells within column C when adjacent cells in column B have matching values. In the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "203-33",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-33/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/203-33/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "204-39",
          "title": "I need a VBA code that can search for the string '0000X000xx1' on Sheet2 of my Excel spreadsheet and change the font col",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "204-39",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/204-39/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/204-39/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "207-25",
          "title": "How do I apply conditional formatting or use a formula to highlight adjacent duplicate names in column A of an Excel she",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "207-25",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/207-25/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/207-25/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "208-20",
          "title": "How can I sort column I, labeled 'R DAYS', in an Excel spreadsheet from highest to lowest using VBA? The spreadsheet I a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "208-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/208-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/208-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "209-30",
          "title": "I have an Excel VBA issue where I'm trying to remove the last three characters from each postcode in a column, and the n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "209-30",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/209-30/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/209-30/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "21-16",
          "title": "I attempted to use a VBA macro to transpose data from a column to a row in Excel for items that have duplicates in colum",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "21-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/21-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/21-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "211-13",
          "title": "I have a column starting at cell A8, extending to the last populated row in the column, that contains numbers with incor",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "211-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/211-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/211-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "212-20",
          "title": "How can I split the text in an Excel cell into separate rows in cells from the B column when the text is separated by tw",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "212-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "212-40",
          "title": "How can I optimize the VBA code I've written to delete duplicate rows based on the ID in Column A of an Excel sheet so t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "212-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "212-45",
          "title": "I need assistance with a macro that can delete entire rows on the active worksheet where column 14 contains the word \"in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "212-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "212-7",
          "title": "I need a macro for Microsoft Excel 2010 that can delete entire rows if any cells within column C contain any of the spec",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "212-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/212-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "22-12",
          "title": "I need a macro for my Excel file that updates the values in column E based on the data in columns A and D. Specifically,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "22-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/22-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/22-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "22-47",
          "title": "I need to sort data in column B based on a helper column (J) so that the first entries written in column I are shown fir",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "22-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/22-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/22-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "2212",
          "title": "How do I create a formula in cell E2 that counts the number of cells within the range $B$2:$B$273 that contain dates gre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "2212",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2212/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2212/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "225-4",
          "title": "I need a VBA macro that will process my data, which is imported with dates in a specific format followed by other charac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "225-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/225-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/225-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "225-40",
          "title": "I need a macro that can go through cells in a column and replace the values of duplicate entries with a predefined value",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "225-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/225-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/225-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "227-40",
          "title": "How can I remove or replace text duplicates within the same row in Excel, as most of the available methods only apply to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "227-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/227-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/227-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "228-19",
          "title": "I have written code intending to delete rows in Column A of an Excel worksheet where each cell's content includes any of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "228-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/228-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/228-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "23-24",
          "title": "How can I automatically delete a word and its associated details in columns B, C, D, and E in my Excel file if the word ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "23-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/23-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/23-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "23-28",
          "title": "I have an Excel workbook with multiple sheets, each sheet containing around 7000 rows of data. I need a macro that will ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "23-28",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/23-28/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/23-28/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "230-16",
          "title": "How can I create a macro in Excel that copies substrings that start from the first letter in each cell of column A and p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "230-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/230-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/230-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "234-30",
          "title": "How can I filter data using multiple criteria that are based on cell references? Specifically, I need to know how to app",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "234-30",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/234-30/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/234-30/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "236-22",
          "title": "I've created a macro to remove blank rows in Column A, but I need to adjust it so that it also deletes the row directly ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "236-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/236-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/236-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "237-21",
          "title": "How can I modify my existing Excel VBA macro which sorts rows starting from row 6 based on the score in Column T, so tha",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "237-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/237-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/237-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "239-15",
          "title": "Is it possible to set up cells in column B of an Excel sheet so that the content of the cells can be modified, but if so",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "239-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/239-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/239-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "239-25",
          "title": "I attempted to create a macro to delete all items in column M of my 'Maturities' sheet where the ageing is less than or ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "239-25",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/239-25/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/239-25/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "24-23",
          "title": "In my Excel spreadsheet, which has over 13,000 rows, I need to filter column A for the groups labeled '@9T', 'SAL', and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "24-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/24-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/24-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "243-30",
          "title": "How can I use VBA to populate a sheet with counts from multiple other sheets using 'countifs' in Excel? In my 'Dev_count",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "243-30",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/243-30/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/243-30/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "247-14",
          "title": "I am working on an Excel task and need to create a macro that calculates frequency dates between two given dates. Specif",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "247-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/247-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/247-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "247-24",
          "title": "I require a VBA code to manage a dataset in an Excel sheet. Specifically, I need the code to dynamically identify and ca",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "247-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/247-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/247-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "248-48",
          "title": "I need help with my VBA code in Excel that's supposed to insert a certain number of rows based on the value in a cell. T",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "248-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/248-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/248-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "249-48",
          "title": "In my Excel sheet, column B contains dates in ascending order and column N contains random numbers. I need to mark an 'X",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "249-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/249-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/249-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "250-20",
          "title": "I need a VBA script that can sum the values in column J of an Excel sheet where the corresponding entries in columns B &",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "250-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/250-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/250-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "253-22",
          "title": "I receive a daily worksheet with over 20,000 rows of data, and I need to condense each row by removing any cells with em",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "253-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/253-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/253-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "254-34",
          "title": "I've provided a spreadsheet with a 'before' and 'after' sheet. In the 'before' sheet, some values are colored green, and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "254-34",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/254-34/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/254-34/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "260-16",
          "title": "How can I move entire rows from the main worksheet to another worksheet when a cell in column 5 contains the value \"No\",",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "260-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/260-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/260-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "262-17",
          "title": "How can I create ranges for header values 'Task' and 'Responsibility' without hard coding the range definitions, and the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "262-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/262-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/262-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "263-1",
          "title": "I have a worksheet with a list that contains columns of material types, widths, and heights, and this list can vary in t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "263-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/263-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/263-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "267-18",
          "title": "I have an Excel sheet where I am concatenating strings in the format [\"/out/\", \"https://theurl.top/images/2.jpg\", \"text?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "267-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/267-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/267-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "267-21",
          "title": "How can I create a comprehensive report in a single sheet ('sheet merging') in Excel that compiles data from multiple sh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "267-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/267-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/267-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "269-43",
          "title": "How can I create a macro in Excel to delete a row if there's a value in column I but the corresponding cell in column H ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "269-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/269-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/269-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "269-44",
          "title": "In my Excel sheet named 'Import Data,' I am trying to create a macro that will delete the four rows located immediately ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "269-44",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/269-44/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/269-44/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "275-49",
          "title": "How do I create A macro in Excel that will merge columns B, C, and D into a single column, with the result referenced to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "275-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/275-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/275-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "2768",
          "title": "In an Excel report concerning outstanding orders, I am looking for a formula that will allow me to assign available inve",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "2768",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2768/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/2768/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "279-23",
          "title": "How can I incorporate a formula that deletes everything outside of parentheses into a VBA macro which iterates over all ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "279-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/279-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/279-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "28-7",
          "title": "How can I delete all rows in an Excel column that occur after and including a part number starting with '400-' followed ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "28-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/28-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/28-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "280-17",
          "title": "How can I remove all rows from an Excel spreadsheet except those with the last unique entry in a specific column? I have",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "280-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/280-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/280-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "281-14",
          "title": "In my monthly Excel workbook, which tracks sales and can include a summary table and sales for each day of the month up ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "281-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/281-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/281-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "282-45",
          "title": "How can I modify my VBA macro to delete all rows in column A from the top of the sheet up to one row before the row wher",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "282-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/282-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/282-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "283-32",
          "title": "I have four sheets in an Excel workbook. In Sheet1, I have a system snapshot, and in Sheet2, I have taken another snapsh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "283-32",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/283-32/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/283-32/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "283-7",
          "title": "I need a VBA Macro to transpose data from the 'DATA' sheet into the 'RESULT' sheet where the amount should be displayed ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "283-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/283-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/283-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "289-3",
          "title": "I need a VBA macro that can remove leading and trailing spaces for each item in cells and reduce any instances of multip",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "289-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/289-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/289-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "290-1",
          "title": "I have a data set that is currently manually transposed in columns K to V, and I would like a macro to automate this pro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "290-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/290-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/290-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "290-27",
          "title": "I am looking for a macro that can be used in Excel 2016 to delete alphabetic characters from cell values in column B, wi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "290-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/290-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/290-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "291-23",
          "title": "How can I create a macro in Excel that performs the action I have manually done on columns J to Q, which involves insert",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "291-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/291-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/291-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "297-42",
          "title": "After running my macro, it only creates a new sheet named 'combined' without pulling in the data from the other sheets. ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "297-42",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/297-42/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/297-42/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "298-22",
          "title": "I have a spreadsheet with clock in times in column C and clock out times in column D. I need a VBA code to calculate the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "298-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/298-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/298-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "30-27",
          "title": "How can I amend my VBA code to delete rows in an Excel sheet where cells in column I from the second row down contain th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "30-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "300-19",
          "title": "How do I remove the percentage symbol and clear the contents of a cell with a value of 0 from a specific column of an Ex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "300-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/300-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/300-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "300-35",
          "title": "I have a list that requires macro automation to insert rows. I have been doing it manually, as exemplified by the two ro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "300-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/300-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/300-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "3002",
          "title": "I need help resolving an issue with my Excel file regarding the Remark column. Originally, I sought help and received a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "3002",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3002/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3002/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "302-1",
          "title": "How can I match and replace adjacent cells in Sheet1 and Sheet2 with those in the ITEMS sheet based on Column B? Specifi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "302-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/302-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/302-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "304-35",
          "title": "How can I delete the values in column F where there are empty cells in column E, as highlighted by the yellow cells in t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "304-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/304-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/304-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "30468",
          "title": "My formula for calculating the number of days in a month, which should be entered into cell L24, is functioning correctl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "30468",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30468/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30468/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "3048",
          "title": "I need assistance with a simple formula to perform a VLOOKUP on data in another table, which is highlighted in yellow. C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "3048",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3048/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3048/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "307-46",
          "title": "I have a column full of data that needs to be split into several columns at the semicolon symbol (;). For instance, cell",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "307-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/307-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/307-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "30709",
          "title": "I have data in cells D6 to D100 of an Excel worksheet, and I need to extract dates from these strings into column E.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "30709",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30709/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30709/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "30930",
          "title": "How can I calculate the number of values greater than zero in Column A that are between the occurrences of the number 1 ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "30930",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30930/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/30930/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31011",
          "title": "How can I create a formula in cell R6 that totals the values within a specified table range where the dates fall between",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31011",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31011/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31011/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31184",
          "title": "I need to adjust the formula in cell H2 so that when the results spill, it doesn't include blank rows displaying zeros. ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31184",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31184/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31184/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "312-46",
          "title": "I need assistance with VBA, as I am a beginner and have been unable to solve my problem. I have a table (Table 1), which",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "312-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/312-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/312-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31202",
          "title": "How can I automatically update the 'debt' column in my payroll table based on the deductions or additions made each week",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31202",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31202/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31202/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31420",
          "title": "I'm trying to use the VLOOKUP function in Excel to match item numbers with their corresponding names, but I encounter an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31420",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31420/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31420/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "315-23",
          "title": "How can I quickly delete rows from an Excel dataset containing over 800,000 rows if a cell in a particular column (appro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "315-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/315-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/315-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31628",
          "title": "How can I determine and extract the number of the last day from a range of dates in cells A1:A11 in my Excel sheet, ensu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31628",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31628/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31628/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31745",
          "title": "How can I convert date formats from 'Day, dd Month yyyy' to 'dd-mm-yyyy' when the data is imported from SAP extracts int",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31745",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31745/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31745/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31746",
          "title": "How do I calculate the total cost savings for a contract given multiple cost savings that could occur in different month",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31746",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31746/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31746/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "319-11",
          "title": "I have a macro in Excel which is supposed to split a cell into separate columns; the cell contains names, addresses, and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "319-11",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/319-11/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/319-11/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "31915",
          "title": "How can I create a formula in Excel that allows me to sum the prices in column B corresponding to each unique serial num",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "31915",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31915/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/31915/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32023",
          "title": "I need a formula to display the current department code for an employee based on today's date. Employees are assigned to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32023",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32023/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32023/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32093",
          "title": "I have an Excel spreadsheet for employee details with column B containing the current employee's name. Columns C, D, and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32093",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32093/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32093/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "321-9",
          "title": "How can I create a VBA code to run in an active workbook named 'Data 1', which can identify and delete paired rows where",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "321-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/321-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/321-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32255",
          "title": "I'm not an expert with spreadsheets, but I have a specific problem I can't figure out. I'm working on a spreadsheet and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32255",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32255/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32255/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32293",
          "title": "How can I create a formula in Excel to populate cells B14 through B17 based on the values in cell A14? The logic should ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32293",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32293/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32293/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32337",
          "title": "I need a formula in Excel that will automatically update the value in a 'CATEGORY' column based on two factors: the date",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32337",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32337/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32337/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32438",
          "title": "How can I format a cell in Excel to only display the time when the current cell contains both the date and the time in t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32438",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32438/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32438/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32468",
          "title": "Can I use a formula to extract data that falls within a 12-month period from a specific starting point? In the example I",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32468",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32468/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32468/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "325-44",
          "title": "I am working with data at a wtype_id, name, and status level, and I need to split the filter data from a column into spe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "325-44",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/325-44/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/325-44/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32562",
          "title": "How can I use COUNTIF in Excel to count occurrences of each unique item number listed in Column C, considering I'm worki",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32562",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32562/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32562/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32612",
          "title": "How can I create a workday column in an Excel file that identifies each date as the day of the week (e.g., Mon, Tue, Wed",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32612",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32612/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32612/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32764",
          "title": "How do I create an Excel formula in Cell D10 that displays a specific value based on the type of runway indicated in Cel",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32764",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32764/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32764/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32789",
          "title": "In my Excel workbook, I have a formula that successfully matches the date value in column A with the primary header in r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32789",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32789/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32789/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32895",
          "title": "I need an Excel formula to calculate the inventory status by using the data from the Sales, Product List, and Purchases ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32895",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32895/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32895/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32902",
          "title": "I need assistance with calculating bonuses in a spreadsheet based on tiered income levels. Specifically, Tier 1 should a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32902",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32902/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32902/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "32975",
          "title": "In my Excel worksheet, column A contains repeated values which define distinct ranges. I'm seeking to calculate and disp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "32975",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32975/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/32975/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "330-23",
          "title": "I need a VB (Visual Basic) code that can sort my Excel data by the Date column. The oldest dates should be at the top, a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "330-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/330-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/330-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33036",
          "title": "I need a formula that will look at the text in cell C2 and output specific values in Column D based on certain text crit",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33036",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33036/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33036/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33094",
          "title": "I've been struggling for the past few days with creating a dashboard in a separate tab of my resource planner. The plann",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33094",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33094/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33094/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33157",
          "title": "I need to know how I can populate column K with the earliest activity date for each individual based on the dates listed",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33157",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33157/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33157/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "333-29",
          "title": "I need to find a specific date using VBA in Excel according to the following conditions: In column P of the 'A' sheet, i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "333-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/333-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/333-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33355",
          "title": "I need a formula that can reference data from one table to another in Excel. Specifically, from the first table, the for",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33355",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33355/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33355/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "334-11",
          "title": "How can I modify my VBA code to delete both blank rows and rows with values between 1 and -1 in Column C of an 'Imported",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "334-11",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/334-11/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/334-11/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33471",
          "title": "I need a formula to convert an amount in Zimbabwean dollars (ZWL) to US dollars (USD) using the exchange rates provided ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33471",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33471/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33471/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "336-1",
          "title": "How can I create VBA code that sorts values in Column A from smallest to largest and places them in Column E (top to bot",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "336-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/336-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/336-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33722",
          "title": "How can I calculate the median market capitalization for a set of companies that fall within a specific percentile range",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33722",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33722/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33722/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33791",
          "title": "How do I use Excel to automatically list all the values from column A corresponding to rows where there's an 'X', given ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33791",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33791/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33791/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "33935",
          "title": " I need help with creating a formula in Excel that will return the latest entry from column J on the condition that the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "33935",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33935/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/33935/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34-5",
          "title": "How can I create a macro to delete rows in an Excel worksheet where the rows contain dates in column L that are earlier ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34-5",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34-5/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34-5/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "340-2",
          "title": "I am looking for a VB (Visual Basic) Code that can process my Payroll data in Excel. The code should iterate through the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "340-2",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/340-2/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/340-2/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34033",
          "title": "How can I create a formula that retrieves data from a column based on whether a given date falls within two adjacent dat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34033",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34033/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34033/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "341-14",
          "title": "I have a dataset in columns A and B, and I need to transform this data to achieve the layout described in the 'result' s",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "341-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/341-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/341-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "341-40",
          "title": "How can I insert a specific text into a new column in an Excel spreadsheet using VBA if three conditions are met in othe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "341-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/341-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/341-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "3413",
          "title": "How can I create a formula in Excel that will sum values when the Department in column A equals 'PI' and the RU in colum",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "3413",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3413/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3413/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34180",
          "title": "How can I return 'True' in column D if there is a partial match between a cell in column B and the cell to its right in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34180",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34180/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34180/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "342-46",
          "title": "How can I create a macro that will extract data from columns B and C on a sheet named 'Imported Data', starting from row",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "342-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/342-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/342-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34210",
          "title": "How can I apply a function or formula in Excel to automatically fill in empty cells in column C with the value in column",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34210",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34210/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34210/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "343-20",
          "title": "I need a macro to select and copy only the duplicated items in column C from the 'Import Data' sheet to the 'Duplicates'",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "343-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/343-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/343-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34365",
          "title": "How can I create a formula in Excel that sums up the annual values from Column B based on dates from Column A, but only ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34365",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34365/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34365/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "344-21",
          "title": "How can I adjust my VBA code to highlight unique values in a column of text strings within Excel, instead of highlightin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "344-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/344-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/344-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34435",
          "title": "I need help with formulating an Excel spreadsheet to calculate Paid Time Off (PTO) based on an employee's start date wit",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34435",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34435/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34435/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34467",
          "title": "How can I identify clients from a spreadsheet of employee visits over the last year who have not been visited by any emp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34467",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34467/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34467/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34469",
          "title": "My formula in cell AH6 is not working to calculate the number of leave days when a month is selected. What should I do t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34469",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34469/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34469/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "34493",
          "title": "The current formula in my Excel sheet's Column O is not producing the intended results. Specifically, I've encountered a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "34493",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34493/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/34493/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "347-49",
          "title": "I need a VB (Visual Basic) Code that can look up data in Column D of my Excel worksheet, identify the text \"Gross Pay\", ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "347-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/347-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/347-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "35207",
          "title": "How can I create a formula in Excel that will calculate a running sum of data numbers found in Column B, grouping these ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "35207",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35207/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35207/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "353-19",
          "title": "I have a spreadsheet with data in the 'MAIN' sheet, and I manually fill in data in the 'ENTER' sheet, where I sometimes ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "353-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "353-29",
          "title": "I have an Excel spreadsheet with 15,000 rows. Each row includes a Unique Reference Number (URN) in column A and associat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "353-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "353-50",
          "title": "How can I set up an Excel macro to accomplish the following tasks? First, it should insert a new column that corresponds",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "353-50",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-50/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-50/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "353-6",
          "title": "I am working with a large Excel sheet, and I need a solution to filter column 'I' to only display 'Hub Notes', then sele",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "353-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/353-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "35387",
          "title": "I need to create a formula in Excel that will count the number of cells in a column that include any errors from a provi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "35387",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35387/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35387/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "355-29",
          "title": "I need a macro for my 'data' sheet in Excel that will format the output in the same way as in my 'Report' sheet. The mac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "355-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/355-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/355-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "35739",
          "title": "How can I correct an error that occurs when calculating a 'Cut Off' time that should be half an hour before the 'STD' ti",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "35739",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35739/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35739/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "35742",
          "title": "I've created a spreadsheet to record the results of an upcoming horse show, which involves multiple sections and aggrega",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "35742",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35742/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35742/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "35747",
          "title": "How can I create a formula in cell M2 that counts the number of values within the range B2:J11 that are greater than 100",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "35747",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35747/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/35747/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "359-21",
          "title": "I created a macro to delete rows in which specific conditions are met in a particular column. The macro operates properl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "359-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/359-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/359-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36097",
          "title": "I need a formula to calculate tax recoupment and scrapping allowance, which should be placed in column I of my Excel she",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36097",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36097/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36097/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36170",
          "title": "How can I extract multiple matching records, such as dates and hours of services provided to clients, from our service r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36170",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36170/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36170/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36191",
          "title": "I am working with data in Column B and I've used the formula =LEFT(B2,FIND(\" \",B2)-1) to extract the first name from a s",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36191",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36191/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36191/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36236",
          "title": "How can I merge the contents of distinct columns into a single cell while eliminating any duplicate values that might ap",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36236",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36236/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36236/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36277",
          "title": "How can I create a lookup formula that returns results based on header names, which works correctly when dragged down in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36277",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36277/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36277/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36281",
          "title": "How can I calculate fuel costs in Table 1 by multiplying the 'pay' and 'gallons' values associated with a particular dat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36281",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36281/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36281/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36478",
          "title": "I need assistance with a formula to match and retrieve cost information from one table to another. In Table 1, I have a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36478",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36478/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36478/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "367-23",
          "title": "I have written a macro that should delete rows from a worksheet named 'Final Recon Items' where the values in column D (",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "367-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/367-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/367-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36745",
          "title": "How can I write a nested IF statement in Excel or an alternative approach to achieve the following outcomes based on val",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36745",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36745/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36745/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36764",
          "title": "I need assistance writing a formula in cell C2 of sheet 1 that sums up the total number of products from sheet 2, range ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36764",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36764/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36764/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "368-44",
          "title": "I need a VB (Visual Basic) Code that will extract the Date, Employee Name, Clock In, and Clock Out times from my employe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "368-44",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/368-44/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/368-44/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "36842",
          "title": "How can I create a formula in Excel to count the number of occurrences of each phrase listed in column A of Sheet2 withi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "36842",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36842/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/36842/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37-23",
          "title": "I need a VB (Visual Basic) code that can look through column D in an Excel sheet for cells that do not contain the text ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37-23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37-23/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37-23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "370-43",
          "title": "I need assistance with writing some VBA code that can insert a row above every line where there is an \"X\" in Column A, s",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "370-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/370-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/370-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37062",
          "title": "I'm new to Excel and need assistance with creating a formula. My worksheet contains three columns: the first for debits,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37062",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37062/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37062/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37086",
          "title": "I need a formula for a column named 'Total' where, if a value in column1 is 'A', it should add 10, but if the value is '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37086",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37086/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37086/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "371-33",
          "title": "I need a VB (Visual Basic) code to delete all rows in my Excel worksheet that contain numerical values. The data in ques",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "371-33",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/371-33/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/371-33/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37228",
          "title": "How can I return the delivery window based on the specified window in Column E, handling stores with either one or two d",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37228",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37228/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37228/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37229",
          "title": "How can I create a formula in Excel that will sum numbers entered in the range A1:G24 based on month end dates entered i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37229",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37229/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37229/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37378",
          "title": "How can I create an Excel formula to place a number in column A, starting at cell A2, that increments by 1 each time a n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37378",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37378/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37378/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "374-18",
          "title": "I attempted to write a macro to delete all rows in Column E of the 'Imported Data' sheet that have values less than 1. H",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "374-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "374-31",
          "title": "How can I modify my existing VBA code to only delete the blank rows that appear above a cell with the text \"Code\" in col",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "374-31",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-31/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-31/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "374-9",
          "title": "I need to create a macro in Excel that searches through row 6 on Sheet 1 for cells containing the letter 'A' and sums th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "374-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/374-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37456",
          "title": "How can I calculate the total sales for each product, for a specific date range between 6:00:00AM and 5:59:59AM of the n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37456",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37456/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37456/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37462",
          "title": "In my Excel file, I've inputted country names and revenues in local currency, which are then automatically converted int",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37462",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37462/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37462/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37554",
          "title": "I'm using a SUMPRODUCT(ISNUMBER(SEARCH formula in cell F2 of my Excel spreadsheet, but it's not functioning correctly. M",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37554",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37554/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37554/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37609",
          "title": "I am working with two columns of data in Excel where each column has values ranging from 1 to 9 and also includes the va",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37609",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37609/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37609/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37615",
          "title": "How can I create an Excel formula in a timesheet that automatically calculates the payment for providers based on partic",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37615",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37615/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37615/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37764",
          "title": "I have a two-column table in Excel with columns for 'Miles' and 'Trip #'. I need a formula that can automatically input ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37764",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37764/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37764/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "379-36",
          "title": "I need a VB Code that will look for a specific text\"Georgia Its Tax\" in column E. If this text is found and the correspo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "379-36",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/379-36/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/379-36/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37900",
          "title": "I need help with a formula in Excel that can automatically return a value based on the current day's date. Also, I have ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37900",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37900/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37900/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "37962",
          "title": "I am trying to modify the value in cell B4 based on the selection made from a drop-down menu in cell A4. While I have a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "37962",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37962/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/37962/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38074",
          "title": "I need to calculate the profit in a table by subtracting expenses, which are scattered across multiple columns, from the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38074",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38074/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38074/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "382-10",
          "title": "I have a column (A) in my Excel spreadsheet, and I would like to delete all values in this column that are not equal to ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "382-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/382-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/382-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "382-29",
          "title": "I have an Excel file with four sheets; three contain data and the fourth is a Totals sheet. I need a macro that can sum ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "382-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/382-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/382-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "383-16",
          "title": "I have a spreadsheet with approximately 90,000 rows that contain duplicate items in column A. I would like to display th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "383-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/383-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/383-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38332",
          "title": "How can I create a formula in Excel to count the occurrence of each unique value in column A only once, regardless of ho",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38332",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38332/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38332/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "384-4",
          "title": "How can I use VBA to determine the differences in values between two data sources in Excel? Please copy values that appe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "384-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/384-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/384-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38462",
          "title": "How can I round up time formatted as Custom [h]:mm, such as '55260:26', to the nearest whole hour in Excel, resulting in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38462",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38462/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38462/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38537",
          "title": "I need a formula that will search through cells B3 to B36 (which could extend to 500 rows) and sequentially number these",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38537",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38537/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38537/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38655",
          "title": "I am attempting to apply the DB (Declining Balance) formula in a larger Excel sheet that includes hundreds of items, eac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38655",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38655/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38655/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38662",
          "title": "I need assistance with using an INDEX and MATCH formula to find a match that corresponds with data in two columns, B and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38662",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38662/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38662/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "387-16",
          "title": "How can I use VBA to clear only one instance of a value in the 'Value' column when it matches a value in the 'Result' co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "387-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/387-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/387-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38703",
          "title": "How can I count the number of unique Item Nos. that have been placed into a Bin and reset this count whenever the items ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38703",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38703/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38703/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "388-47",
          "title": "I am attempting to write VBA code to delete all rows in a worksheet where the items beginning with 'Z1' or 'X1' appear i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "388-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/388-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/388-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38823",
          "title": "I need a formula to sum values in a range between two dates where a cell contains a specific search term. The values to ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38823",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38823/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38823/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38969",
          "title": "How can I create a single formula in column R that will produce results based on multiple if conditions specified in my ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38969",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38969/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38969/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "38985",
          "title": "I'm dealing with two tables in Excel. In table 1, I have a column with duplicate names, and in table 2, I have those sam",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "38985",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38985/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/38985/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39046",
          "title": "How can I use a formula in Excel to check if a reference in column B is marked as 'CANCELLED' in another column, then co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39046",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39046/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39046/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "3911",
          "title": "How can I distribute 120 different currency notes among employees based on their salaries using Excel? I have provided a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "3911",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3911/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/3911/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39180",
          "title": "How can I look up and sum values in Excel where my lookup array is A3:A57 and B3:B57, and I want to sum the correspondin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39180",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39180/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39180/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39190",
          "title": "How can I delete all the rows in Excel that contain a specific value, such as '108', which is found in column 'A'? I am ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39190",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39190/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39190/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39272",
          "title": "How can I remove the numbers and the text that appear before ':' from each row in my Excel spreadsheet? Currently, rows ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39272",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39272/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39272/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39432",
          "title": "How do I create an automated formula in Excel for column M to calculate the per unit cost of an item based on FIFO metho",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39432",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39432/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39432/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39490",
          "title": "How can I round the fourth digit of the value in a cell that is highlighted in yellow, rounding up if the digit is 5 or ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39490",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39490/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39490/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "395-36",
          "title": "I have three sheets: 'Main unique ID' for manual input of IDs that may change, 'Result what I am getting', and 'Result w",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "395-36",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/395-36/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/395-36/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "395-48",
          "title": "How can I create a macro in Excel that highlights cells in one column incrementally where the cumulative sum exceeds the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "395-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/395-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/395-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39515",
          "title": "How can I modify my Excel formula to return a value based on both month and year, rather than just month? I have a works",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39515",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39515/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39515/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39604",
          "title": "In my Excel file, add one more condition to my current formula: If the value of B4 is less than or equal to 20, then the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39604",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39604/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39604/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39667",
          "title": "How can I enter a 'description' in a table below a specific (black) row and automatically receive the corresponding 'Pay",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "398-14",
          "title": "I have data across multiple sheets where I need to match and sum values from columns B, C, and D that are repeated acros",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "398-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "398-22",
          "title": "How can I write a macro to remove all duplicates of text items in the same row of Column I that contain 'Fuel,' except f",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "398-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "398-47",
          "title": "I need a VB (Visual Basic) code that can delete rows in an Excel spreadsheet which contain duplicate values in Column O.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "398-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/398-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "399-14",
          "title": "I need a VB (Visual Basic) Code to split recorded employee hours into regular hours and overtime hours. The hours worked",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "399-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/399-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/399-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "399-8",
          "title": "In my Excel sheet, can I use VBA code to delete every row that contains a 'False' value in any cell, and then copy and p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "399-8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/399-8/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/399-8/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39903",
          "title": "How can I count individual bin locations that do not start with 'X' or 'Z' in a single cell field, where each location i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39903",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39903/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39903/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39931",
          "title": "I need to return a value from a third column by looking up and matching data contained within two other specific columns",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39931",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39931/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39931/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "39946",
          "title": "I need a formula that calculates the revenue recognition per invoice, where the period is specified for each invoice. Ho",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "39946",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39946/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/39946/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40-43",
          "title": "I need a macro that can count occurrences of a number or text within a range based on three specific filters: date, trac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "402-43",
          "title": "How can I duplicate rows in a spreadsheet based on a count specified in column AG, so that if the value in AG is 1, the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "402-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/402-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/402-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40234",
          "title": "How can I create a formula in Excel that will display all relevant email addresses in cell K5 when a specific company is",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40234",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40234/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40234/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "404-14",
          "title": "How can I reduce the dataset for each individual name by deleting any row that contains a zero value in column I and als",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "404-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/404-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/404-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40467",
          "title": "When I add a day to a date in Excel, for instance taking the date 12/31/2023 in cell A1 and adding one day to it in cell",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40467",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40467/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40467/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40478",
          "title": "How do I use a formula to extract text from column A in a spreadsheet, starting after the second space character to the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40478",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40478/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40478/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40510",
          "title": "How do I calculate the interest value using a formula that takes into account multiple criteria from other columns in my",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40510",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40510/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40510/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40536",
          "title": "I need to calculate a value in cell C3 of my Excel worksheet by multiplying the number of non-blank cells within the ran",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40536",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40536/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40536/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40757",
          "title": "How can I use Excel to sum the values in the second column of a spreadsheet when those values correspond to a specific m",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40757",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40757/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40757/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40758",
          "title": "How can I create a formula in Excel to remove all commas from the data in Column A and display the resulting text withou",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40758",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40758/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40758/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "408-39",
          "title": "I need to shift the entire column labeled '0-15' to Column B in Excel by inserting it as a new column, but the location ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "408-39",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/408-39/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/408-39/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "408-5",
          "title": "How can I create a macro in Excel to calculate the sum of values in Column D for rows where Column E contains a zero, in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "408-5",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/408-5/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/408-5/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40809",
          "title": " In my workbook, I have an existing formula within the range of cells F9:F19, and I need assistance on how to modify it.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40809",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40809/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40809/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40892",
          "title": "I need a formula to extract the color mentioned in each cell of column 1 and display that same color in column 2. For ex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40892",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40892/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40892/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "409-45",
          "title": "I need a macro that deletes empty rows based on the value of cell I2. The macro should have two conditions. First, it sh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "409-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/409-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/409-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40943",
          "title": "How can I count the number of non-blank items in Column B that are associated with a specific account number, 6327008, l",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40943",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40943/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40943/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "40959",
          "title": "Is there a way to create a formula in Excel that acts like an 'If UNLESS' condition? Specifically, I have a spreadsheet ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "40959",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40959/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/40959/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41-12",
          "title": "Is there a macro that can count the number of names in the range F7:F12 that are formatted in bold and display the count",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41-47",
          "title": "How can I split account names into multiple ranges across different sheets based on the information in column B, sum the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41265",
          "title": "What I am looking for: \n   Exact matching each root word (D2-D5) to all in column A (A2-A6) and adding the search volume",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41265",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41265/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41265/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41297",
          "title": "How can I populate data in Column A to correspond with the daily data in Column B by matching the day, month, and year i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41297",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41297/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41297/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "413-15",
          "title": "How can I transpose data from a sheet named 'archive' horizontally into another sheet named 'output' while retaining the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "413-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/413-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/413-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41348",
          "title": "In my Excel file containing project data, each project, identified by a consistent budget ID in column H, may have multi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41348",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41348/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41348/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "414-20",
          "title": "How can I create a macro in Excel to delete all rows above the first occurrence of 'Invoice No.' in column A?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "414-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/414-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/414-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41410",
          "title": "I have a table in Excel with data across various arrays that includes information on heating, cooling, lighting, and oth",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41410",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41410/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41410/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41420",
          "title": "How can I achieve the result shown in the C2:D8, where strings are deduplicated and sorted in ascending order? The conte",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41420",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41420/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41420/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "415-45",
          "title": "I am learning how to use a dictionary in Excel VBA and am encountering a problem.  I need assistance with understanding ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "415-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/415-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/415-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41575",
          "title": "How can I create a formula in Excel that counts as 1 each time a value in cell B2 is less than or equal to 65% of the va",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41575",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41575/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41575/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41589",
          "title": "How can I create a function that populates a cell in Column J with a specific action based on the last contact date in C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41589",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41589/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41589/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "416-15",
          "title": "I have an Excel spreadsheet with a column of dates where some cells include a time value that starts with 'T' and ends w",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "416-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/416-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/416-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "416-27",
          "title": "I have written a macro intended to delete all blank rows in Column A, starting from the  row thatthe first value in that",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "416-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/416-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/416-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41601",
          "title": "How can I create a system in Excel to automatically reference the correct sheet based on a list of student names in Colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41601",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41601/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41601/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41680",
          "title": "How can I remove an unwanted space at the beginning of cell content in a column with approximately 5000 rows in my Excel",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41680",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41680/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41680/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41691",
          "title": "Why am I getting an error with my VLOOKUP formula in Excel, and how can I correctly use it to display 'two' in cell F11 ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41691",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41691/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41691/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41692",
          "title": "I'm utilizing a solution from this forum, but I still encounter an issue. When performing a lookup in my Excel file, I n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41692",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41692/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41692/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "418-25",
          "title": "I need a macro that can add '-Fuel' at the end of the text present in column I for each row where the corresponding cell",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "418-25",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/418-25/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/418-25/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "418-29",
          "title": "How can I write a VBA macro for Excel that processes a range where each cell in column A contains a reference number and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "418-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/418-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/418-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41801",
          "title": "How can I compare the data in column E of the 'Records1' tab with the data in column E of another tab named 'Records2' a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41801",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41801/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41801/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41857",
          "title": "How can I calculate the minimum distance, in meters, between a specific point—defined by its UTM north and east coordina",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41857",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41857/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41857/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41937",
          "title": "I am attempting to create a simple visual summary, much like a catalog overview, from a large dataset. My main challenge",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41937",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41937/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41937/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41969",
          "title": "How can I apply the formula '=COUNTBLANK(A3:C3)' to multiple cells in a pattern where the cell range shifts by three col",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41969",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41969/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41969/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "41978",
          "title": "How can I create a formula in Excel that counts the number of open cases for each year? Specifically, I need the formula",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "41978",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41978/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/41978/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42034",
          "title": "How can I use Excel functions to sum a range of values in Column H based on corresponding index-matched values from Colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42034",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42034/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42034/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42058",
          "title": "How can I create a SUM formula in an Excel table with KPIs for each month that automatically adjusts to sum the values u",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42058",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42058/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42058/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42110",
          "title": "How can I create a formula that extracts unique values from a table in Sheet1 with the range A1 to T16 and then places t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42110",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42110/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42110/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42181",
          "title": "How can I modify my SUMIF function to sum values in column B only if the corresponding cells in the reference column I c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42181",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42181/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42181/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42198",
          "title": "I need a formula for cell C2 which will calculate the following logic: if all instances of \"Potato\" in column A have a c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42198",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42198/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42198/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42216",
          "title": "How can I transform the dataset from columns (Group A) to rows (Group B) in Excel, excluding values such as -9999 and 'M",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42216",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42216/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42216/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42217",
          "title": "How can I create a formula in Excel for infilling data in one group (Group C) using data from two other groups, where Gr",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42217",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42217/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42217/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "4225",
          "title": "In an Excel worksheet, I have two columns. Column A contains text strings, each including a color name such as 'Green', ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "4225",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4225/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4225/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "423-16",
          "title": "How can I use VBA to selectively transpose cells from one sheet to another in Excel? Specifically, I want to transpose c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "423-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/423-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/423-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42354",
          "title": "I need a formula for Excel that evaluates cells in a row across columns A, B, and C, and returns the result in column D.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42354",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42354/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42354/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42378",
          "title": "How can I apply a formula in Excel that will return a value of 1 only when a cell in column C is blank and its adjacent ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42378",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42378/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42378/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42515",
          "title": "How can I modify an Excel formula in a foreign exchange rate table (TABLE 2) so that when I drag down the formula it upd",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42515",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42515/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42515/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42526",
          "title": "How do I assign points to a range of data values where the points should vary proportionately between -5 and +5, such th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42526",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42526/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42526/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42566",
          "title": "I have data where I want to map results from one column to another based on the match of multiple keywords. The setup is",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42566",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42566/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42566/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "427-35",
          "title": "I need help with a VBA macro in Excel that can remove duplicates based on two specific conditions: duplicated identifier",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "427-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/427-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/427-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42798",
          "title": "How can I use the Index/Match formula or an alternative method in Excel to list out item numbers corresponding to specif",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42798",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42798/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42798/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42897",
          "title": "How can I devise an Excel formula to calculate the amount of time that has elapsed between two timestamps listed in Colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42897",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42897/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42897/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42902",
          "title": "How can I convert a list of over 5,000 rows, where every 3rd row contains an address, into a format where the address ro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42902",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42902/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42902/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "42930",
          "title": "How can I assign a unique number to each subdivision based on the earliest house built within it, to sort a longer list ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "42930",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42930/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/42930/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "430-1",
          "title": "How can I create a report in Excel where my original data spans from columns A to H and I want to generate a complete da",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "430-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "430-39",
          "title": "How can I write a macro to transpose columns F:O into columns B:E in Excel, with specific columns transposed into specif",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "430-39",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-39/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-39/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "430-43",
          "title": "How can I create a macro that allows me to input a key variable in a 'Filter Criteria' tab to filter the data across all",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "430-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/430-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43026",
          "title": "I'm struggling with a SUMPRODUCT formula I've written as '=SUMPRODUCT((Table!A5:A17=summary!A10)*(Table!B5:B17=summary!D",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43026",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43026/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43026/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43213",
          "title": "How can I calculate the average of a changing range in Excel, where the range increases as new data is added, and have t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43213",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43213/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43213/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43227",
          "title": "How can I automatically fill in corresponding values from one chart to another in Excel, based on matching criteria? The",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43227",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43227/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43227/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "433-47",
          "title": "How can I add to or subtract from cells across five unit columns (from unit 1 to unit 5 for addition, and from unit 5 to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "433-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/433-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/433-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "434-7",
          "title": "How do I write a VBA code that can take a value from a merged cell in column A, such as the value 36354.4 in a merged ce",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "434-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/434-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/434-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43406",
          "title": "How can I automatically number headings and subheadings in Excel where users identify them as 'H' for heading or 'S' for",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43406",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43406/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43406/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43436",
          "title": "How can I calculate the number of records that are still open by the end of each month, given two dates for each record ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43436",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43436/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43436/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "435-36",
          "title": "I am regularly updating a badge report in Excel and need to automate the process to save time. The updated data includes",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "435-36",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/435-36/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/435-36/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "435-38",
          "title": "How can I create a macro in Excel that highlights cells in one column incrementally where the cumulative sum exceeds the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "435-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/435-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/435-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43589",
          "title": "How can I calculate the number of days using Excel if I have a range of dates in a cell formatted as text, such as '2 to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43589",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43589/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43589/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43657",
          "title": "When I specify a person's name, I want to know how many times that name is highlighted in green in excel\nThe current for",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43657",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43657/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43657/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43695",
          "title": "How can I extract the filename and its extension (such as .png or .jpg) from a set of URL links where the text to be ext",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43695",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43695/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43695/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43775",
          "title": "How can I copy a list of products into every third cell of a block of four across a specified range in Excel?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43775",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43775/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43775/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "438-18",
          "title": "How can I replace specific values in an array with new values while copying all data from a 'sheet INPUT' to a 'sheet OU",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "438-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/438-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/438-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "438-24",
          "title": "How can I create a macro in Excel that identifies and deletes rows starting from a duplicated 'Stock Number' header and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "438-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/438-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/438-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "43939",
          "title": "How can I import data from Sheet2 to Sheet1 in Excel by matching the values in cell B1 of both sheets and then filling t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "43939",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43939/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/43939/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "440-24",
          "title": "How can I automatically replace the data in the same cell with expanded values in Excel, so that the values in column B ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "440-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/440-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/440-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44017",
          "title": "How can I create a formula in Excel that calculates a new rate based on an 'annual, quarterly, or semi-annually' frequen",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44017",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44017/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44017/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "441-38",
          "title": "I need to know how to automatically count the occurrences of red and green values within a single cell in Excel.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "441-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/441-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/441-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44264",
          "title": "How can I create a dynamic range formula in Excel to calculate percentage payments over a moving 3-week slot, which auto",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44264",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44264/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44264/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44266",
          "title": "How can I use the INDEX and MATCH functions to retrieve data from a specific cell in the 'Data' tab and place it into a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44266",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44266/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44266/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44278",
          "title": "I have two columns in Excel, with the first column containing street names and cities and the second column containing z",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44278",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44278/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44278/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44287",
          "title": "How can I create a formula in Excel that calculates a rolling sum of values from another column over the course of a mon",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44287",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44287/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44287/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44296",
          "title": "How can I convert data from a range in Sheet 1 to a column format in Sheet 2 by matching units with states using a formu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44296",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44296/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44296/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44344",
          "title": "How can I create a formula in Excel to sum up multiple employee shifts, assigning a numerical value of 1 to each 'Full d",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44344",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44344/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44344/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44389",
          "title": "How can I create a formula in Excel that returns the column header of the smallest value greater than zero in each row, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44389",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44389/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44389/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "444-14",
          "title": "How can I create a macro that extracts data from two different tabs, titled 'Data' and 'Basic Master' in an Excel workbo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "444-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/444-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/444-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "444-27",
          "title": "I need a macro that will copy data from the 'Imported Data' sheet to the 'Extract' sheet on 'Sheet2'. The condition is t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "444-27",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/444-27/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/444-27/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44628",
          "title": "How can I group and sum the total sales of all products by using the first letter of each product's name, such as 'A' fr",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44628",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44628/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44628/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "448-11",
          "title": "I found a code in this forum that highlights and deletes duplicate rows within the current region in Excel, using the en",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "448-11",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/448-11/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/448-11/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "44913",
          "title": "I have an excel table, each table has the corresponding value of different ID in different months (suppose it is the tur",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "44913",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44913/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/44913/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "450-9",
          "title": "I am looking for a macro that would insert a row and label 1, 2, 3, etc in column C above the name of a business that wo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "450-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/450-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/450-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45025",
          "title": "How can I create a formula in Excel to extract the portion of data following the second set of numbers within a cell?   ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45025",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45025/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45025/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45030",
          "title": "How can I find the closest value that is 'At Least' a certain cost within a tier range using an Excel formula, without g",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45030",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45030/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45030/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45063",
          "title": "How can I use a formula in Excel to populate a list of currently out-of-stock products with the quantities sold last yea",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45063",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45063/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45063/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45181",
          "title": "How can I create a list that shows the number of days left for each student to return their library books, using their S",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45181",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45181/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45181/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45194",
          "title": "How can I optimize a complex Excel formula that involves IF statements and lookups, to handle large datasets (over 70K r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45194",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45194/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45194/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "453-49",
          "title": "How can I split a quantity listed on Sheet1 into batches based on available inventory by batch on Sheet2, using the firs",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "453-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/453-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/453-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45300",
          "title": "How can I delete all text from a point where a specific word ('FT') appears within strings in an Excel column? Each row ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45300",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45300/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45300/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45372",
          "title": "How can I calculate a forecast total value in Excel column E based on the current time? For instance, before 09:45, the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45372",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45372/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45372/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "455-35",
          "title": "How can I modify my VBA code to delete rows in an Excel worksheet where the first character in the cells of column B is ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "455-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/455-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/455-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45581",
          "title": "How can I modify a formula in Excel to fill a vacant position in a string.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45581",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45581/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45581/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "456-19",
          "title": "How can I sort a named range of four-digit item numbers situated in a non-standard grid layout in Excel, such as in cell",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "456-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/456-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/456-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "4562",
          "title": "Is there a straightforward method for using COUNTIF with multiple ranges? I'm asking this because I've been applying col",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "4562",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4562/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4562/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45635",
          "title": "I need a formula for an Excel sheet that assigns a score of 6, 3, 2, or 0 in rows 5, 6, and 7 across columns B, C, D, an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45635",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45635/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45635/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45707",
          "title": "How can I create a formula in Excel that counts the number of occurrences of a specific value (e.g., 'x') in the past mo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45707",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45707/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45707/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45738",
          "title": "I am attempting to automate the creation of cashflows for a list of approximately 1,500 bonds in my database, avoiding t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45738",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45738/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45738/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "458-24",
          "title": "In column F of my worksheet, I have cells with text where I need to delete all content before and including the first sl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "458-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/458-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/458-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "458-49",
          "title": "I need a macro that can delete rows from the second row onwards in my Excel worksheet where the text in Column C ends wi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "458-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/458-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/458-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45896",
          "title": "In my Excel sheet, I'm struggling to display multiple dates in a single cell in Column I using a date format rather than",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45896",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45896/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45896/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45937",
          "title": "How can I create a formula in Excel that automatically calculates the percentage category based on a range of values in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45937",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45937/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45937/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "45944",
          "title": "I'm having issues with a nested IF statement that should compare the names and start dates between a table and data loca",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "45944",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45944/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/45944/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46058",
          "title": "How do I create a formula in Excel that appends an 'S' from cells in column B to the corresponding cells in column A and",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46058",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46058/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46058/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46121",
          "title": "How do I configure Excel to read a date (month) on one sheet and then place the total amount spent in that month under t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46121",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46121/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46121/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46167",
          "title": "How can I create a formula in Excel that will calculate the sum of multiple cells subject to a specific criterion? Speci",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46167",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46167/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46167/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "462-45",
          "title": "I need a macro to extract all rows from the 'Imported Data' sheet into an 'Extract' sheet, where the sum of values in Co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "462-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/462-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/462-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46240",
          "title": "How can I create a formula in Excel where cell J displays 'Yes' if cell H either contains a date or the word 'Yes', disp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46240",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46240/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46240/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "463-17",
          "title": "I have a list of company names in column A and their respective counts in column B of an Excel sheet. I need to generate",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "463-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/463-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/463-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46444",
          "title": "I am using an INDEX MATCH formula in Excel that needs to look for a job code and a day code to return the task start tim",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46444",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46444/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46444/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46495",
          "title": "How can I calculate the sum of amounts based on the month without adding an extra column in Excel? I have a list of date",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46495",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46495/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46495/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46646",
          "title": "How can I calculate the percentage of night shifts for a given month, considering a variable year range from 2010 to 202",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46646",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46646/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46646/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46856",
          "title": "Is there a method to sum values in Excel when working with two sets of criteria, where one criterion is based on a verti",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46856",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46856/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46856/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "46897",
          "title": "How can I create a formula in Excel that counts the instances of a specific text or number in a column that matches a sp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "46897",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46897/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/46897/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "469-9",
          "title": "I need a macro that will take debit and credit values from Column C in my Excel spreadsheet and separate them into two n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "469-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/469-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/469-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "470-22",
          "title": "I need to amend my code to extract account numbers from the 'Acc Numbers' sheet to the 'Extract Acc Non Zero' sheet, spe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "470-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/470-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/470-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "4714",
          "title": "I need help with calculating a moving average for each employee listed in column A of my spreadsheet, specifically on a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "4714",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4714/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/4714/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "472-15",
          "title": "I need a VB (Visual Basic) Code that can perform the following actions: If cell A1 contains the text '4Ozark', then set ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "472-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/472-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/472-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47353",
          "title": "I am employed at a school and we want to analyze how grades in various subjects correlate with attendance in those class",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47353",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47353/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47353/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47360",
          "title": "How can I modify my Excel formula so that it doesn't count blank cells as a match when checking if values in A2 and B2 f",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47360",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47360/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47360/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47484",
          "title": "How can I create a formula in Excel that generates a sequential number in column C, starting from 01, based on the corre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47484",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47484/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47484/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47699",
          "title": "In my Excel sheet, I perform vlookups that result in 'TRUE' or 'FALSE' values. I need a way to automatically calculate t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47699",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47699/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47699/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "477-40",
          "title": "How can I write a macro that combines the names from the first two columns in Excel to create a new name in the third co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "477-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/477-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/477-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "477-45",
          "title": "How can I create my Excel VBA code to efficiently merge and sum duplicate values by taking into consideration columns A,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "477-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/477-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/477-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47741",
          "title": "How can I modify a formula using the Workday function in Excel to return a range of dates as short dates instead of seri",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47741",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47741/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47741/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47766",
          "title": "I am attempting to calculate my agent's yearly production which includes yearly bonuses that reset annually. In the Exce",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47766",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47766/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47766/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47798",
          "title": "How can I extract the collection name from cells where the 'Collection-' tag is placed at different positions within the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47798",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47798/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47798/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47827",
          "title": "How to assemble three numbers into a three-digit number, with the numbers arranged in order from largest to smallest.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47827",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47827/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47827/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47842",
          "title": "I have a VBA routine that standardizes job descriptions by updating words in the past tense to the simple present tense.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47842",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47842/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47842/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "47933",
          "title": "How do I correct my IF formula in Excel so that in cell E2 it looks up the value in the hours column (B2), and multiplie",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "47933",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47933/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/47933/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48034",
          "title": "How can I calculate the average of data that is in a 'Years/Months' format (e.g., 7/5 for 7 years and 5 months) in Excel",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48034",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48034/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48034/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48080",
          "title": "How can I create a column in Excel that displays the value of every nth row from another column? I've tried using OFFSET",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48080",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48080/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48080/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "482-22",
          "title": "I need some assistance in further refining my VBA code for an Excel sheet. Previously, with the help of forum members, m",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "482-22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/482-22/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/482-22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48257",
          "title": "How can I extract numerical values and determine the valid decimal sign from a text string containing various currency p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48257",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48257/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48257/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48354",
          "title": "How can I look up text in a column in Excel and return one of three specific values based on the content of the cell? I ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48354",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48354/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48354/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48357",
          "title": "How can I calculate the 'Remaining OH' quantities based on 'Total OH' and 'Future Usage' for parts that repeat in a colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48357",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48357/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48357/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48365",
          "title": "How can I sum amounts from a data sheet using SUMIFS when a user can select up to 3 regions for a product, with an 'All'",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48365",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48365/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48365/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48378",
          "title": "How can I calculate the difference between two receipt numbers based on a specific Batch number, and display the result ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48378",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48378/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48378/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48527",
          "title": "How can I retrieve unique text from the third column of a worksheet where the first two columns contain the same text, a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48527",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48527/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48527/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48588",
          "title": "In my Excel file, I have a list of different products in Column A with 15 unique records, but these records are not sort",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48588",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48588/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48588/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "486-17",
          "title": "I have a list of numbers in cells that represent dates, formatted as 'jjjjmmdd' without spaces (e.g., 20210305 for March",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "486-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/486-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/486-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48608",
          "title": "I'm attempting to generate a report in which an agent's name appears alongside their 'session ID' whenever that session ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48608",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48608/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48608/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48620",
          "title": "I have a list with products where some are identical but have different serial numbers. For instance, in cell A1, I have",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48620",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48620/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48620/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48643",
          "title": "I am trying to create an Excel spreadsheet to track the total accumulated wealth over a period of time with variable dep",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48643",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48643/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48643/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48685",
          "title": "I need a formula to retrieve and add the value from the last non-zero cell in Column K to the current cell value in Colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48685",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48685/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48685/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48715",
          "title": "I need to create a COUNTIF formula in cell D1 that will count the number of cells in the range A1:A7 where the first let",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48715",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48715/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48715/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48735",
          "title": "How can I find rows in Sheet1 when the alphanumeric data in column A and the dates in column B match corresponding data ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48735",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48735/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48735/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48745",
          "title": "How can I look up multiple product codes that are located within the same cell, separated by semicolons, in an Excel tab",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48745",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48745/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48745/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48799",
          "title": "How can I modify my nested IF statements in Excel to sum up the results of multiple VLOOKUP functions applied to product",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48799",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48799/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48799/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "488-14",
          "title": "How can I sort a column in Excel using VBA to organize the data based on the frequency of occurrence of each item instea",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "488-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/488-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/488-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "488-29",
          "title": "In my Excel sheet, I have a situation where if the first occurrence of a value in column A matches the first occurrence ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "488-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/488-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/488-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48921",
          "title": "How can I create an Excel formula that removes the leading characters from a variable-length number, where the specific ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48921",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48921/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48921/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48924",
          "title": "How can I list all the missing dates from a sequence in column A and place them into column C in Excel?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48924",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48924/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48924/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48930",
          "title": "I am trying to calculate the Kindergarten year for students in our preschool program. To determine eligibility for kinde",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48930",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48930/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48930/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48950",
          "title": "I am attempting to use a formula in cell B2 that will determine the week number based on the date given in column A. The",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48950",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48950/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48950/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48968",
          "title": "I am struggling to remove the text 'Garments manufactured For Winnerforce' from a cell only when it appears in isolation",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48968",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48968/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48968/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48969",
          "title": "How can I compare and match the columns and rows of two separate data sources in Excel to get a result, as demonstrated ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48969",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48969/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48969/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48975",
          "title": "How can I copy values from columns 'A' and 'B' in one row from the 'Input' worksheet to the 'Output' worksheet if the co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48975",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48975/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48975/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48982",
          "title": "Why does the formula max.number in cell B323 not return the maximum+1 but shows 000, and why does it result in a circula",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48982",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48982/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48982/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "48983",
          "title": "I'm struggling to transfer data from one table to another using Excel. I attempted to use the INDEX/MATCH method but it ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "48983",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48983/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/48983/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "490-49",
          "title": "How can I identify blocks of cells in an Excel spreadsheet by finding a specific value (designated 'Identity') in column",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "490-49",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/490-49/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/490-49/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49036",
          "title": "How can I calculate and display the percentage in cell B8 of the 'dashboard' sheet by dividing the numerical value in ce",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49036",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49036/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49036/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "491-6",
          "title": "How can I merge cells in columns B, C, and D to correspond with merged cells in column A of my Excel dataset, such that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "491-6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/491-6/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/491-6/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49118",
          "title": "How can I modify my Excel formula to return the minimum value in column C for each matching range in column A, similar t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49118",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49118/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49118/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49196",
          "title": "How can I modify the formula in column C of my Excel sheet so that it updates the year to 2022 in the date when the year",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49196",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49196/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49196/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49237",
          "title": "How can I add a period to a cell in Excel if there is a middle initial present, and leave it blank if there is no middle",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49237",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49237/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49237/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49246",
          "title": "How can I create a formula in Excel that allows me to look up and display a restaurant's name based on an outlet number,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49246",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49246/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49246/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49278",
          "title": "I have a deferred income spreadsheet where I need the formula in column K to return a zero if both the subscription star",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49278",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49278/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49278/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "493-18",
          "title": "How do I match the corresponding data in columns A and F, and then delete the data that does not appear in column F in t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "493-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/493-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/493-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "493-5",
          "title": "How do I create a macro in Excel that will delete rows in which the value in Column C matches the value in Column D and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "493-5",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/493-5/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/493-5/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49300",
          "title": "I'm attempting to retrieve a value from a table in Excel using two criteria with INDEX and MATCH functions, but my formu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49300",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49300/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49300/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49333",
          "title": "I'm encountering an issue where my VLOOKUP formula returns #N/A, even though I've attempted to troubleshoot common probl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49333",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49333/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49333/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "494-13",
          "title": "How can I use VBA code in Excel to automatically highlight non-matching pairs of values in two different columns, and al",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "494-13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/494-13/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/494-13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49455",
          "title": "How can I construct a formula in cell H3 that will count the number of times a range of values (K1:K9) appears in column",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49455",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49455/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49455/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49467",
          "title": "How can I convert a series of times formatted as 'XXXX AM' or 'XXXX PM' without colons into military time for the purpos",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49467",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49467/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49467/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49490",
          "title": "In my Excel file, I have a vlookup formula that I am trying to apply within the 'PRICE LIST' sheet, specifically ranging",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49490",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49490/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49490/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "495-31",
          "title": "How can I use VBA to transpose data from a 'Data' sheet to a 'Result' sheet as demonstrated in an attached Excel file? S",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "495-31",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/495-31/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/495-31/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "495-40",
          "title": "How can I create a macro in Excel that edits cells in columns L and N, starting at row 3 down to the last row with data ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "495-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/495-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/495-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "496-15",
          "title": "How can I write VBA code to loop through all data in column E of my Excel worksheet, identify cells containing the text ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "496-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/496-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/496-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "496-34",
          "title": "I need assistance with an Excel macro to handle duplicate ID numbers in column A. Specifically, I want to count the tota",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "496-34",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/496-34/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/496-34/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49613",
          "title": "How do I write a formula in cell DATA U2 that performs different sum calculations based on the condition of cell E2? If ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49613",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49613/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49613/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49654",
          "title": "I need a formula to check for blank cells in a range (A2:D2) and return a specific text indicating which columns are bla",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49654",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49654/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49654/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49665",
          "title": "How can I write an IFS formula in Excel that examines different columns (D, E, F, G, H) in a row to return a specific 'S",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49665",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49665/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49665/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49667",
          "title": "How can I write formulas in Excel to populate the start and finish times of meetings, which are represented in columns F",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49709",
          "title": "How do I modify my existing IF function in Excel to display a blank cell if there is no date in column K before it evalu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49709",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49709/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49709/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49782",
          "title": "How can I create an automatic numbering function in Excel that generates a new change order number in a sequence? I need",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49782",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49782/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49782/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49801",
          "title": "In Excel, how can I combine the results of both the LEFT 2 characters of a cell and RIGHT 3 characters of another cell i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49801",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49801/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49801/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49834",
          "title": "How can I enter a comic book name into cell A1 and then search across all comic book names in Row 1 to find and list the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49834",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49834/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49834/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49857",
          "title": "I'm working on a spreadsheet and would like to create a sequence of numbers in column G that starts with 1 and continues",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49857",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49857/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49857/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49859",
          "title": "How can I create a formula that will allocate a series of month numbers in column C across a range of period columns (D ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49859",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49859/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49859/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "499-10",
          "title": "I'm in the process of learning VBA and currently focusing on the For Each Loop. I want to understand whether it's possib",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "499-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/499-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/499-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49939",
          "title": "How can I modify the formula '=SUMPRODUCT(($A$4:$A$13=$A20)*($B$4:$B$13=B$19)*COUNTIF($D$17,Data[[#Headers],[31-03-2021]",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49939",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49939/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49939/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "49945",
          "title": "How can I create an Excel formula that will allow me to sum the quantities in range D4:D4 associated with a specific mod",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "49945",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49945/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/49945/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "5-33",
          "title": "How can I count 'AD' sets in Excel 2010, considering only the two rows above and below a given cell, highlight the cell ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "5-33",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5-33/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5-33/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50051",
          "title": "In my Excel worksheet, I have encountered an issue where the ID# column CC is intended to reflect line numbers matching ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50051",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50051/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50051/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50066",
          "title": "How can I create a second sheet in Excel where I can search for a person's name and then have their information from the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50066",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50066/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50066/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50088",
          "title": "I am currently faced with the problem of not being able to count items in an adjacent cell if the cell in a particular c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50088",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50088/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50088/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50154",
          "title": "How can I create a search function in Excel where typing the name of a comic book title into a search box displays the c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50154",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50154/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50154/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50179",
          "title": "How can I identify fruits like 'ORANGE' which are sold simultaneously in four different stores using data from the provi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50179",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50179/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50179/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50193",
          "title": "How can I modify my SUMPRODUCT formula in Excel to get a ranking that does not include duplicates? Currently, when two p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50193",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50193/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50193/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50232",
          "title": "I need to track rest days for NHL teams during the 2021-2022 season using an Excel spreadsheet I've attached. Specifical",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50232",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50232/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50232/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50233",
          "title": "I need a formula to allocate daily work orders to a varying number of team members, ensuring an even distribution. The n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50233",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50233/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50233/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50250",
          "title": "How do I use the value in cell G2 ('sela') as A reference to sum the numbers of the values corresponding to G2 in column",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50250",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50250/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50250/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50322",
          "title": "I need help using the index/match function in Excel to retrieve and sum the values for each month from the raw data on t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50322",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50322/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50322/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50324",
          "title": "How can I determine the percentage of occurrences where two products were sold together, as shown in the two tables in m",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50324",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50324/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50324/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50329",
          "title": "I have two columns with dates and need to set up a formula in Column J with multiple IF conditions to manage the followi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50329",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50329/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50329/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "504-10",
          "title": "As someone new to Regular Expressions, I need help creating a macro that uses RegExp to extract data from a sheet where ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "504-10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/504-10/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/504-10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50442",
          "title": "How can I modify the Excel formula for calculating Mark-to-Market in trading, specifically for cells G17, G18, and G19 i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50442",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50442/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50442/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50452",
          "title": "I am working on creating a grid for a duty assignment schedule that places an 'x' in cells to indicate a match between n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50452",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50452/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50452/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50472",
          "title": "How can I extract only the number from the string in the provided Excel file?",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50472",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50472/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50472/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50486",
          "title": "How can I configure column A in Excel to display the text 'Chep' or 'Loscam' based on which of the two corresponding cel",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50486",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50486/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50486/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50521",
          "title": "I need a formula to count the number of blank cells between the first and second numeric entries across each row of a ta",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50521",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50521/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50521/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50526",
          "title": "I need a formula that I can place in cell B9 of my spreadsheet and then drag down to dynamically look up and display val",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50526",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50526/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50526/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50534",
          "title": "I am trying to combine two specific formulas into a single cell in Excel for a work schedule that calculates weekly hour",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50534",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50534/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50534/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50573",
          "title": "I am trying to run a report twice at different times at my work and then copy and paste the results into a new Excel fil",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50573",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50573/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50573/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50577",
          "title": "I have a downloaded file containing a column with the header 'Job Title,' and each cell in this column contains text. I ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50577",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50577/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50577/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50630",
          "title": "How can I split a sentence into three parts using 'ADDED' and 'DUE' as common words that indicate where to divide the se",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50630",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50630/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50630/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50631",
          "title": "I'm trying to create a formula in Excel to return the workdays that fall between the dates in cells B3 and F3, but I nee",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50631",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50631/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50631/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50641",
          "title": "I am using the formula '=IFERROR(C2/B2,0)-1' to calculate the week-on-week difference in an Excel spreadsheet. However, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50641",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50641/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50641/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50681",
          "title": "I need a formula to check if a value in a column appears more than once. For the first occurrence of the value, the outp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50681",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50681/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50681/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50682",
          "title": " I am currently stuck on my spread sheet trying to find the sum of data in multiple columns. I'm working on a spreadshee",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50682",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50682/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50682/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50683",
          "title": "As a high school golf coach who tracks player scores, I can calculate a player's season average easily. However, I'm str",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50683",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50683/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50683/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50755",
          "title": "How do I calculate the difference between the second and third values in each column of my Excel workbook? For example, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50755",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50755/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50755/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50768",
          "title": "How can I automatically plot each risk ID from a Risk Assessment table, which includes an Impact and a Likelihood number",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50768",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50768/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50768/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50796",
          "title": "I am seeking a more efficient way to handle increasing preorder products in Excel as manual counting has become too time",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50796",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50796/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50796/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50811",
          "title": "How do I use the indexmatch formula to look up a percentage that is tied to a specific sales representative and their cl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50811",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50811/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50811/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50816",
          "title": "How can I retrieve the row number of the next highest number in a column in an Excel spreadsheet when the starting value",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50816",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50816/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50816/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50916",
          "title": "How can I create an Excel school calendar where typing a cycle day number into a column automatically populates the clas",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50916",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50916/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50916/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50952",
          "title": "How can I retrieve the corresponding ID from a table in Excel based on multiple search parameters? The parameters are fo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50952",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50952/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50952/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "50971",
          "title": "How can I extract values from column B that are associated with duplicate entries in column A? Also, in my spreadsheet, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "50971",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50971/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/50971/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51-12",
          "title": "How can I create a VBA code that will count the number of times a value in a range from A1 to M1 on an Excel sheet incre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "510-3",
          "title": "How can I calculate the total number of days from a date provided in column D of an Excel sheet, where the date is in ye",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "510-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/510-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/510-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51030",
          "title": "How can I use XLOOKUP in cell D14 to search within two tables where one set of dates (in the top table) changes daily, a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51030",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51030/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51030/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51090",
          "title": "How can I calculate the difference between the total cartons received as indicated in Column M and the errors present in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51090",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51090/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51090/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51249",
          "title": "How do I create an Excel formula that outputs different results based on the text values of two different cells where if",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51249",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51249/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51249/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51262",
          "title": "How can I use the SUMIFS function in Excel with criteria that come specifically from a list or range I have specified, t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51262",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51262/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51262/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51289",
          "title": "In my spreadsheet consisting of a range from A1:H2, the first row contains only the text 'cat' or 'dog' while the second",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51289",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51289/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51289/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51313",
          "title": "How can I adjust my INDEX MATCH formula in Excel to list all names corresponding to a particular status (e.g., 'A' for a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51313",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51313/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51313/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51354",
          "title": "How can I extract the month and year from the last 6 characters of text strings in column A, when there isn't always a '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51354",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51354/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51354/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51359",
          "title": "How can I calculate the total of all consecutive 1's in a column in Excel, where only the groups of adjacent cells with ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51359",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51359/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51359/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51431",
          "title": "Is there a formula that can help me extract the first name from a string of names in a cell? I'm aware of the TEXT TO CO",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51431",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51431/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51431/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51556",
          "title": "I need to create a formula that will calculate the average market rent for a specific storage unit size and improvement ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51556",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51556/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51556/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51586",
          "title": "I need a formula to check if the text from one cell is found within another cell that contains additional text. For exam",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51586",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51586/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51586/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "516-46",
          "title": "I need a macro that extracts data from a range beginning at cell A2, extending across multiple columns to column E, and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "516-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/516-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/516-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51661",
          "title": "How do I create a formula in cell J3 that assigns a value based on the value in cell H3, such that if H3 is equal to or ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51661",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51661/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51661/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51680",
          "title": "How can I identify which areas of business use an item that's in stock by analyzing a spreadsheet with columns (A to F) ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51680",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51680/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51680/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51741",
          "title": "How can I generate a sequence of continuous numbers in column C for the IDs listed in column A, taking into account the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51741",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51741/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51741/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51807",
          "title": "I have a situation where I want to convert quantities into cartons and remaining boxes in Excel. For example, if I have ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51807",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51807/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51807/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51811",
          "title": "I need a formula that can scan a range B2:Y2 in a sheet called HERE and return a value into column H2 on a sheet named D",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51811",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51811/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51811/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51859",
          "title": "How can I calculate the total hours each employee has spent on each category by summing the values from column H when co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51859",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51859/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51859/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51863",
          "title": "How can I set up an Excel cell under the 'Next Review' column to automatically calculate a date that is 7 days after the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51863",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51863/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51863/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51894",
          "title": "How can I create a dynamic SUMIFS formula in Excel that adjusts the criteria range based on the current month, consideri",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51894",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51894/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51894/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51899",
          "title": "How can I return the row number (in range) of the LAST occurrence of the MIN value of a range of numbers that satisfy a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51899",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51899/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51899/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51913",
          "title": "How can I extract a value from a cell in a column that contains a longer phrase incorporating the text 'XYZ12345'? For i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51913",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51913/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51913/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "5192",
          "title": "How can I create an Excel formula to reformat data in column A such that the last occurring number in each cell is moved",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "5192",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5192/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5192/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "51958",
          "title": "How can I convert a date and time from a format such as '9th Aug 10:08' to a more standardized form like '09/08/21 10:08",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "51958",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51958/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/51958/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "520-31",
          "title": "I attempted to create a VBA code to remove everything from row 3 except for the cells in Column C that contain the phras",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "520-31",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/520-31/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/520-31/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52021",
          "title": "In our daily statement form, we have a dropdown list in the 'particulars' column. How can we set it up so that when a pa",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52021",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52021/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52021/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52050",
          "title": "I require help to streamline an existing Excel formula. My current method calculates the total achievements expected in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52050",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52050/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52050/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52216",
          "title": "I'm having some trouble with getting my INDEX-MATCH formula to return values from subsequent rows beneath the first row ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52216",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52216/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52216/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52220",
          "title": "How can I create a formula in Excel that will display the sum of two horizontal cells in a third cell, only if both of t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52220",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52220/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52220/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52233",
          "title": "How can I modify my Excel formula so that when I drag it down, the column reference changes (e.g., B3:B5 to C3:C5), but ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52233",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52233/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52233/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52292",
          "title": "How can I create a formula in Excel that adjusts the referenced cell downwards when the formula is copied across to diff",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52292",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52292/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52292/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52305",
          "title": "How can I create a table in Excel, similar to a heat map, that will count entries based on four criteria: account name, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52305",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52305/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52305/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52379",
          "title": "I need to find out, for each phone number that has called more than once between January 1st and January 7th as document",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52379",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52379/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52379/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "524-2",
          "title": "I've developed a macro to delete duplicates in an Excel worksheet based on a condition, but it's running too slowly. The",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "524-2",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/524-2/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/524-2/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "524-31",
          "title": "I have a list of transaction descriptions in a format referred to as 'bank speak', which includes transaction identifier",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "524-31",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/524-31/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/524-31/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52450",
          "title": "I require a formula for cell M5 that will calculate the number of days remaining from today until the target date specif",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52450",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52450/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52450/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52532",
          "title": "How can I automatically calculate the number of wins for each player in my spreadsheet which currently tracks game score",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52532",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52532/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52532/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52541",
          "title": "I need a formula for the remarks column (J) in my Excel spreadsheet for tracking accounts receivable, because the curren",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52541",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52541/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52541/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52575",
          "title": "How can I use Excel to count the number of weeks year-to-date in which an employee, identified by an employee number, ha",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52575",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52575/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52575/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52633",
          "title": "I need a formula for my Excel sheet that checks if a value in column D appears more than once when coupled with a branch",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52633",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52633/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52633/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52640",
          "title": "I need to filter out all the rows in Column G of the 'RAW' tab in my Excel sheet where the value is greater than 0 and r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52640",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52640/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52640/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52658",
          "title": "I am working with a dataset in Excel where columns A-M contain random numbers that can be positive, negative, or zero. C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52658",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52658/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52658/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52763",
          "title": "How can I update 'sheet2' with data from 'all data sheet' based on a serial number I enter in 'sheet2' using Excel, cons",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52763",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52763/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52763/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "528-20",
          "title": "I have data in Excel starting from A3 to K6, followed by 5 rows of data that I don't need (from Row 7 to Row 11), and th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "528-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/528-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/528-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52807",
          "title": "In my Excel workbook, I am currently using a SUMIFS formula on the 'Data' tab to total my 'actuals' within the range E6 ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52807",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52807/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52807/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52841",
          "title": "How can I use an IF formula in Excel to display different data depending on the TRUE or FALSE status of a checkbox? For ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52841",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52841/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52841/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52888",
          "title": "How can I write a formula in Excel for a forecast document where the revenue calculated by multiplying average order val",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52888",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52888/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52888/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52917",
          "title": "I need to create a formula that can combine day, month, and year data from separate columns A, B, and C into a complete ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52917",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52917/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52917/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52931",
          "title": "I need to extract a value from a 'Data' sheet in Excel based on a date found on the 'Calcs' sheet. Specifically, I want ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52931",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52931/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52931/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "52964",
          "title": "I need to create a formula for cell C2 that locates the number from A2 within column Q, and where a match is found, the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "52964",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52964/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/52964/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53-12",
          "title": "How can I clear the contents of columns B and C in the 'Interest' worksheet if the count of non-empty cells in column B ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53-14",
          "title": "In my Excel workbook, I have values in Column B and corresponding months and years in Column C. Whenever there's a case ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "530-28",
          "title": "How can I find the unique order numbers from column A in an Excel spreadsheet where each order number corresponds to a '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "530-28",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/530-28/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/530-28/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53023",
          "title": "How do I modify my XLOOKUP function in Excel to work with wildcard characters when I'm trying to reference two separate ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53023",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53023/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53023/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53062",
          "title": "I need help creating a formula in Excel for column Q that returns '1' under two conditions: The text in the same row's c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53062",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53062/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53062/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53068",
          "title": "In my Excel worksheet, I need to automate the pay calculation by choosing between two different formulas based on user i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53068",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53068/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53068/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "531-18",
          "title": "I need to transpose data from the 'Data' tab into the format shown on the 'Transpose' tab of my Excel file, specifically",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "531-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/531-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/531-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "531-48",
          "title": "I wrote an Excel VBA macro that was supposed to filter the data by three criteria, but when I tried to run the code, I r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "531-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/531-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/531-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53117",
          "title": "I need a formula for column D in my Excel worksheet that works as follows: if the value in column C is not 'US', then th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53117",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53117/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53117/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53161",
          "title": "I need to calculate the total sum of different categories for each time interval, where the total is determined by which",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53161",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53161/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53161/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53167",
          "title": "I am having difficulty calculating the amount of product used since its last usage. In my attached sample sheet, columns",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53167",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53167/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53167/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53174",
          "title": "I need help creating a formula that will look up a projected completion date from a 'Task_table' and return the correspo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53174",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53174/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53174/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "532-24",
          "title": "How can I retrieve the values from 'Sheet 2' in an Excel workbook if I have a list of corresponding cell addresses locat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "532-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/532-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/532-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "532-3",
          "title": "In the Excel workbooks I have, which contain 25K to 50K lines each, there are two sheets. The first sheet contains the o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "532-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/532-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/532-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53280",
          "title": "How can I create a formula in Excel that will extract the text in column A that appears after the second backslash and b",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53280",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53280/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53280/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53305",
          "title": "I need to sum all prices in cell K3, where it should meet several criteria: 1) Match the 'Part' column range A2 to A4 wi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53305",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53305/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53305/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53367",
          "title": "How can I adjust my Excel formula to compare the value in cell A3 only to non-blank cells in the range A4 to A20 without",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53367",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53367/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53367/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53383",
          "title": "How can I create a formula in Excel to compare names and corresponding statuses across two worksheets, indicating 'match",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53383",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53383/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53383/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "534-26",
          "title": "How can I write a macro that will delete the entire column in an Excel sheet if the Date value in the column is earlier ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "534-26",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/534-26/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/534-26/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "534-40",
          "title": "How can I apply an AutoFilter in Excel to the 'ProjectID' column of the 'headline_tbl' using an array of 'Active' projec",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "534-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/534-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/534-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53449",
          "title": "How can I categorize patient visits from a dataset into separate categories such as Office Visits Only, Covid Vaccines O",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53449",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53449/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53449/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "535-20",
          "title": "How can I delete all columns in an Excel worksheet that contain the keyword '/description' in the first row, given that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "535-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/535-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/535-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53533",
          "title": "How can I automatically fill a profit or loss statement table with respective GL codes that update when I change the mon",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53533",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53533/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53533/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "536-37",
          "title": "I have an Excel worksheet with over 7000 rows of data in column 'A'. Each cell in column 'A' contains several items sepa",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "536-37",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/536-37/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/536-37/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53636",
          "title": "I need assistance determining how many times a number appears within a cell, or alternatively, how many plus signs are p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53636",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53636/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53636/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53647",
          "title": "I need to create a formula where it calculates based on the input of either cell C7 or D7. A number will be input into e",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53647",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53647/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53647/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53860",
          "title": "I need to create a dynamic formula in Excel that can take a forecasted volume entered on a monthly basis for a span of 1",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53860",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53860/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53860/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53957",
          "title": "How can I count the number of times 'x' appears under each category in a table column, where the categories are in separ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53957",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53957/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53957/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53965",
          "title": "How can I set up an Excel sheet so that when I enter a size category (such as Small, Medium, Large, Huge, Gargantuan) in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53965",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53965/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53965/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "53994",
          "title": "I'm trying to create a summary matrix in Excel, where I have an array that includes data columns, each with a date range",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "53994",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53994/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/53994/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54057",
          "title": "How can I create a formula in Excel that automatically enters 'Pass' or 'Fail' in column D based on test results inputte",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54057",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54057/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54057/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54085",
          "title": "How can I create a formula in Excel that populates column C with all the items from column B, with the exception of excl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54085",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54085/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54085/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54105",
          "title": "Can I combine an extract text function with trim and clean functions in Excel? I'm currently using the formula '=RIGHT(C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54105",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54105/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54105/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54144",
          "title": "How can I create an Excel 365 formula to sum the quantities of trucks listed on the 'Data' Sheet when the truck names ar",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54144",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54144/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54144/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54196",
          "title": "I'm struggling with a VLOOKUP function that seems to be failing. I aim to use this function to retrieve the lead time fo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54196",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54196/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54196/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54201",
          "title": "I have a spreadsheet with two rows, each containing unique numbers ranging from 1 to 10. I would like to combine these t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54201",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54201/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54201/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54205",
          "title": "I am working with strings representing 'ACCT-PROG-LOC' that I want to combine into a single string in various ways. The ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54205",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54205/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54205/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54238",
          "title": "I need a formula for cell F8 that matches the amount in a cell on the 'DATABANK SHEET' based on the month and year enter",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54238",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54238/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54238/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54242",
          "title": "How can I create a formula for column E in Excel that will display 'Revenue' in each cell if the corresponding cell in t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54242",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54242/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54242/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54274",
          "title": "I'm new to Excel and need assistance with creating a nested IF formula for a problem I'm working on. Although I've attac",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54274",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54274/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54274/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54337",
          "title": "How can I create a SUMPRODUCT formula that incorporates multiple criteria and also utilizes the SIGN function within an ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54337",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54337/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54337/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54474",
          "title": "I am working with a file that contains a tab named WHP and another with WHP data. I need to collect specific information",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54474",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54474/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54474/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54490",
          "title": "I have been using the AVERAGEIFS formula to calculate averages in my table, specifically intended to exclude cells that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54490",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54490/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54490/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "545-35",
          "title": "I receive data from an external source that includes cells with alphanumeric and wildcard character contents. These cell",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "545-35",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/545-35/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/545-35/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54513",
          "title": "How can I calculate and display the price of an item after applying a 55% discount to its original price of $34.99 in Ex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54513",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54513/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54513/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54590",
          "title": "How can I create a formula in cell GK29 to calculate the sum across columns GK and GR based on cells in column GO being ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54590",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54590/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54590/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54638",
          "title": "How can I use a non-array formula to select and list only the unique values from a range of cells, specifically from A2:",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54638",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54638/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54638/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54640",
          "title": "I'm having trouble getting the DATEDIF function to work in my Excel spreadsheet. The formula I'm using is =DATEDIF(C2,TO",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54640",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54640/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54640/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54667",
          "title": "How can I modify the VLOOKUP formula in range G3:G58 to return a time value from the corresponding row in range M3:M58 i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54675",
          "title": "How can I modify my Excel formula so it counts all occurrences of the name 'Gary Stoddard' in column B, including those ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54675",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54675/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54675/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "547-18",
          "title": "I need a macro that can identify rows in Column A of the 'Imported Data' sheet where the cell content begins with two as",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "547-18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/547-18/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/547-18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "547-43",
          "title": "How do I transpose a selected column from the 'Comp1-performance' range in the 'Emp' table to a row, while keeping the p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "547-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/547-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/547-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54717",
          "title": "How do I use Excel to sum the values in a specific column based on particular start and end row numbers, where the colum",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54717",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54717/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54717/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54747",
          "title": "How can I create a formula in Excel to find and return each person's booked holiday dates? I have a holiday planner grid",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54747",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54747/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54747/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54781",
          "title": "How can I use COUNTIFS to return the number of cells in column A that each have a corresponding cell in column B with a ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54781",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54781/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54781/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "54925",
          "title": "I'm using an array formula to retrieve results in column B where H2:H40 matches the value in M3 and K2:K40 matches the v",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "54925",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54925/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/54925/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55039",
          "title": "How can I create a formula that manages withdrawals from an account balance?   The first withdrawal should be $20,000 wh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55039",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55039/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55039/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55049",
          "title": "How can I create a formula in Excel that allows me to sum values across three columns based on a single criterion? I've ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55049",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55049/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55049/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55060",
          "title": "How do I create a formula in Microsoft Excel that will make cell J23 exactly replicate the value selected in cell I12, b",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55060",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55060/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55060/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55085",
          "title": "How can I use a formula to perform a multiple criteria lookup in the 'Result' worksheet, specifically in row 10?  Is the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55085",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55085/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55085/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55126",
          "title": "How can I create a formula in Excel that will display a result from Column C when I input any one piece of data from eit",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55126",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55126/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55126/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55135",
          "title": "How can I retrieve the gas price value from my 'Data' worksheet and have it returned to cell I2 based on multiple criter",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55135",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55135/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55135/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55224",
          "title": "SpreadsheetBench V1 full benchmark task 667",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55224",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55224/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55224/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55258",
          "title": "How can I create a singular formula to add up values in a column which contain a specific text, for example, '100 *', by",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55258",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55258/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55258/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55260",
          "title": "I need to round the amount in the second column to the nearest $5. I wonder if there is a formula that can do this in Ex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55260",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55260/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55260/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55392",
          "title": "In my Excel worksheet named 'Bank Balance', I have a cumulative formula in column AL that I want to stop calculating aft",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55392",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55392/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55392/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55399",
          "title": "How do I create a formula in Excel that returns 'OK' if a number in one cell falls within a specified tolerance range in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55399",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55399/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55399/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55421",
          "title": "How can I create a formula in Excel that will check the numbers in Column A and return specific text in Column F based o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55421",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55421/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55421/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55427",
          "title": "I am attempting to consolidate data about schools from multiple spreadsheets using INDEX and MATCH functions. Specifical",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55427",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55427/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55427/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55457",
          "title": "SpreadsheetBench V1 full benchmark task 674",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55457",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55457/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55457/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55468",
          "title": "How can I modify my existing INDEX/MATCH formula to incorporate four matching criteria (two vertical and two horizontal)",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55468",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55468/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55468/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55515",
          "title": "In my Excel workbook, I have two tabs. On the first tab, named 'Main', I need to count specific occurrences of an ID, fo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55515",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55515/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55515/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55535",
          "title": "How can I automatically update the text in a cell to create a sequential numbering system that reflects a change in year",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55535",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55535/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55535/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55572",
          "title": "How can I modify an Excel formula to display each row of data for a specific code selected from a dropdown list, startin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55572",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55572/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55572/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "556-37",
          "title": "I have a VBA code that generates new tabs for each unique month listed in a column and copies all corresponding rows to ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "556-37",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/556-37/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/556-37/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55609",
          "title": "How can I create an Excel formula that compares the numbers in the second row with those in the eleventh row, and then, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55609",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55609/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55609/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55610",
          "title": "How can I calculate the duration in hours between two times across different days, for instance, from 20:00 hours on Mar",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55610",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55610/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55610/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55667",
          "title": "I need help creating a formula that takes a letter from cell B40 and finds it within a range spanning from columns E to ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55708",
          "title": "How can I use the AVERAGEIFS function in Excel to calculate the average turnaround time for a given department when the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55708",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55708/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55708/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55794",
          "title": "I have an Excel table with the status of various projects and I want to extract and display the status of these projects",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55794",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55794/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55794/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55817",
          "title": "I need to figure out how to take symmetrical data that is currently arranged in one column and display it in a specific ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55817",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55817/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55817/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55834",
          "title": "How can I create an Excel formula that makes a cell in column K return 'NO' if the corresponding cells in columns D or H",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55834",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55834/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55834/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55860",
          "title": "I need an Excel formula to match data across columns F, G, and H with the data provided in a worksheet; the result shoul",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55860",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55860/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55860/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55877",
          "title": "SpreadsheetBench V1 full benchmark task 688",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55877",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55877/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55877/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55883",
          "title": "I am working on a time and attendance spreadsheet where I need to log both the hours worked by each individual and the n",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55883",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55883/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55883/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55912",
          "title": "In the Test.xlsx file attached, I need to count the number of times the word 'Yes' appears in column AC on the 'Data' sh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55912",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55912/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55912/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55931",
          "title": "I need to know how to compare the chart positions of the Top 50 songs on a week-by-week basis.   In my spreadsheet, colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55931",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55931/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55931/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55965",
          "title": "How can I display the last 10 results from a data sheet on Sheet1 from columns G to P, based on the comparison between t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55965",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55965/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55965/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55976",
          "title": "How can I tally the 'Yes' responses from an 80-question survey in Excel, where each question requires users to mark an '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55976",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55976/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55976/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55977",
          "title": "I'm using the WEEKDAY function to find out the day of the week from a date, with the intention for Monday to be represen",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55977",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55977/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55977/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "55979",
          "title": "How can I create a formula in Excel that will display the name of a supplier, which corresponds to a specific sheet, in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "55979",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55979/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/55979/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56-46",
          "title": "I need a macro that can delete a specified range of cells if there is no match found between two columns. Specifically, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "560-12",
          "title": "How can I efficiently count the quantity of each item from a table that sometimes contains over ten thousand rows of dat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "560-12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/560-12/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/560-12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "560-16",
          "title": "How can I write a VBA code in Excel that takes a list of products and their sales amounts from columns A and B, combines",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "560-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/560-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/560-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56225",
          "title": "How can I display information in an Excel spreadsheet that consists of the date, mechanic, and the total time per day? F",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56225",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56225/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56225/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56274",
          "title": "I need an Excel formula that will automatically populate cells D9, D10, D11, and D12 with the Opening Balance, Debits, C",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56274",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56274/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56274/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56378",
          "title": "How can I create a formula in Excel that will copy products from a list in frame 1 into frame 2, but only include the pr",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56378",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56378/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56378/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56419",
          "title": "How can I create a single formula in Excel to pull a list of non-zero values from 'Quantity' and return the associated T",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56419",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56419/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56419/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56427",
          "title": "How can I automate the transposition of data in Excel where the transposition of values in column G should begin at a ce",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56427",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56427/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56427/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56451",
          "title": "I'm using an Index Match function in Excel to pull in values from another sheet and want to copy the formula across mult",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56451",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56451/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56451/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56475",
          "title": "How can I create a formula in Excel to go through a column O in which 'y' means a number in the corresponding cell of an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56475",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56475/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56475/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "565-19",
          "title": "I need help creating VBA code for Excel that filters one column based on a value in another column, specifically display",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "565-19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/565-19/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/565-19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56561",
          "title": "In my Excel spreadsheet, I'm attempting to retrieve the last recorded value in Column H, which I'm referring to as 'cumu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56561",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56561/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56561/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56563",
          "title": "How can I create a formula in Excel to sum a range of cells that contains the #NUM error value, as illustrated in the at",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56563",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56563/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56563/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56576",
          "title": "How can I modify the formula used in Column C, which averages values from Columns D through J, to automatically exclude ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56576",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56576/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56576/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56597",
          "title": "How can I use Excel to reorganize cells containing comma-separated data where each cell can have sequences prefixed by '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56597",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56597/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56597/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56599",
          "title": "How can I find the Book Number associated with a given slip number on my Excel sheet where each book contains 10 slips, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56599",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56599/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56599/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "566-40",
          "title": "I am trying to enhance a macro for my work project to automate the process of deleting all columns in a sheet where the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "566-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/566-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/566-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56637",
          "title": "I'm working on automating a staffing worksheet to reduce manual input and I need to automate the population of shift num",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56637",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56637/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56637/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "567-21",
          "title": "I need assistance with creating a macro for a worksheet containing over 100,000 rows of data. The macro should compare a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "567-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/567-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/567-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56786",
          "title": "How can I create a rolling average in Excel with a dynamic range based on the previous 365 days for a data set that does",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56786",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56786/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56786/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56855",
          "title": "How can I adjust the formula `=RIGHT(A3;5)-LEFT(A3;5)` to calculate the hours between two times in a single Excel cell A",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56855",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56855/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56855/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56915",
          "title": "How can I populate columns C, D, and E on the COVER sheet with the correct data from the DATA sheet, where the entity is",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56915",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56915/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56915/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56920",
          "title": "I need a formula to extract 10-digit numbers or letters from a text string in Excel where one sequence always starts wit",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56920",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56920/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56920/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56921",
          "title": "I am struggling to get a SUMPRODUCT formula to work correctly in Excel. The goal is to sum every 4 columns, with the cri",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56921",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56921/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56921/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56953",
          "title": "How do I create a formula in Excel that allows me to input a Power (kW) value in cell F2 and then display the correspond",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56953",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56953/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56953/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56975",
          "title": "How can I perform a reverse VLOOKUP in Excel where the wildcard is applied to the table array rather than the lookup val",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56975",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56975/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56975/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56995",
          "title": "How can I create a formula in Excel that operates based on the number of quizzes attended and the number of errors made ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56995",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56995/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56995/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "56996",
          "title": "I need a formula to be placed in column K for the PRICE, that corresponds to a selection I make from a dropdown menu in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "56996",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56996/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/56996/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "570-43",
          "title": "How can I automate the process of extracting specific data from a body of text copied into Sheet1 and inserting it into ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "570-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/570-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/570-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57026",
          "title": "How can I create a single formula in Excel to display only certain digits from a number that is formatted as text with a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57026",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57026/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57026/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57029",
          "title": "I need a formula that will return 'True' in cell C1 if cell A1 contains any text from cell B1. For instance, if A1 has t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57029",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57029/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57029/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57033",
          "title": "How can I use a formula to perform a 3-way match in cell L2 of Sheet 4 in my Excel workbook? I've attempted to use an IN",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57033",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57033/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57033/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57072",
          "title": "How can I modify the XLOOKUP function in Excel so that it skips certain values returned based on additional criteria? Sp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57072",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57072/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57072/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57090",
          "title": "I need a formula that can scan through a column of project numbers and determine the number of unique projects completed",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57090",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57090/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57090/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57105",
          "title": "How do I use the SUMIFS function in Excel to calculate the total amount of money received each month, when the amounts a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57105",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57105/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57105/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57113",
          "title": "I am working with Twitter data and attempting to create a new Excel column that extracts all hashtags from each tweet. W",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57113",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57113/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57113/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57117",
          "title": "I need to create an Excel formula that looks up text within a cell and returns a specific text result in another cell. F",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57117",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57117/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57117/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57137",
          "title": "How can I determine a result in Excel when, for a particular date, an item count that appears more than once is matched ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57137",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57137/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57137/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57225",
          "title": "How can I create a dependent dropdown list in Excel where selecting a name from the dropdown list in cell B1 displays co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57225",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57225/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57225/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57232",
          "title": "How can I retrieve the corresponding volume from a workbook if the origin port, equipment size, and destination port mat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57232",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57232/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57232/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57262",
          "title": "I own a small independent business and am trying to track order volumes, expenses, and profits. My aim is to automate th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57262",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57262/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57262/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57275",
          "title": "How can I find and return up to two matching cell values between a dynamic range of 1-10 lines in a table and a named ra",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57275",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57275/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57275/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57317",
          "title": "How can I set up a function in Excel that would allow me to input a product name and subsequently display the names of a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57317",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57317/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57317/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57354",
          "title": "I am seeking a formula for cell B25 in my Excel sheet that would identify all columns in row 24 containing the word 'sal",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57354",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57354/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57354/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57376",
          "title": "I'm using the formula '=RANDBETWEEN(E3+14,E3+RAND()*100)' in column 'F' to generate an END date that should range from 1",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57376",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57376/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57376/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57427",
          "title": "I am looking for a formula, preferably non-array, that will allow me to extract values from column B of a sheet named 'H",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57427",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57427/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57427/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57445",
          "title": "How can I create a formula in Column E on my Pricing Worksheet that will reference manually inputted data in Columns C a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57445",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57445/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57445/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57448",
          "title": "How can I create a formula in Excel that retrieves the column heading (highlighted in yellow) from a matrix of pipe wall",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57448",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57448/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57448/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "575-15",
          "title": "I have a table with data in the range C10:C100 and I need to replace any entry starting with 'LF ' with the standalone w",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "575-15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-15/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "575-16",
          "title": "I have a large amount of Excel data arranged vertically with student codes and their corresponding class codes. I need a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "575-16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-16/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "575-25",
          "title": "How can I write a VBA code to compare dates formatted as text across columns D, K, S, and Y, and input the latest of the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "575-25",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-25/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/575-25/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57558",
          "title": "I need to create a formula that retrieves a commission rate based on multiple criteria which include a salesperson, type",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57558",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57558/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57558/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57590",
          "title": "I need to calculate a sum in cell B26, where the sum is based on the values in Column L but only includes those values w",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57590",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57590/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57590/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57612",
          "title": "In my Excel worksheet, I have a formula in cell D4 linked to cell C4 which needs to be modified. I want to display diffe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57612",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57612/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57612/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57662",
          "title": "How can I automatically fill in rows highlighted in orange with the corresponding names from my dataset when the rows ha",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57662",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57662/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57662/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57693",
          "title": "I need a formula to calculate the total number of each alphabet in a single cell containing a string of text. For exampl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57693",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57693/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57693/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "577-40",
          "title": "How can I delete entire rows in an Excel worksheet if both Column J and Column K are either blank or contain any of the ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "577-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/577-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/577-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57716",
          "title": "How can I create a custom calendar in Excel where each cell contains both numbers and text, but I want a formula that wi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57716",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57716/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57716/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57728",
          "title": "B1 to G1 have aggregate formulas that change depending on the filtering in column H,When I filter Column H by colour, I ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57728",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57728/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57728/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57743",
          "title": "In my Excel spreadsheet, I'm struggling to create a formula that will update a price book. I have model numbers listed i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57743",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57743/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57743/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57771",
          "title": "How do I determine in Excel that if a value from one row array matches a reference cell value, then the result should be",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57771",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57771/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57771/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57783",
          "title": "I need to sum a row of data if another row shows Cap. The SUMIF formula works for that but now I need to sum the data if",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57783",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57783/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57783/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57866",
          "title": "How do I create a formula in Excel where, if a value in column C on Sheet2 matches a value in column A on Sheet1, the co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57866",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57866/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57866/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57928",
          "title": "How can I automatically remove the outdated article numbers from each cell in a column of product descriptions in an Exc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57928",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57928/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57928/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "57989",
          "title": "In an Excel file that lists the trip location per driver and per day, I'm looking to create a synthesis table that shows",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "57989",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57989/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/57989/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58032",
          "title": "How do I create a formula in cell A2 to search through columns I:K for a person's name based on a combination of their T",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58032",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58032/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58032/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58047",
          "title": "How can I convert the data format in Column G of the sales report generated by my company's system to a new format as re",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58047",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58047/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58047/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "581-46",
          "title": "I have an Excel workbook containing multiple sheets with randomly named sheets. Each sheet has the same format but diffe",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "581-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/581-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/581-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58109",
          "title": "How can I create a formula in Excel to sum the values of the previous four workdays, considering that my current formula",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58109",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58109/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58109/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58114",
          "title": "I need a formula that can extract a number along with subsequent values or text from a string in column I, but only if a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58114",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58114/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58114/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58138",
          "title": "Is there any way, formula to figure out, what is the latets graduation ( column \"D\") according to the Year ( column \"C\")",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58138",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58138/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58138/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58147",
          "title": "I am seeking a formula that will allow me to extract only the values or text from column I and display the desired outco",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58147",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58147/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58147/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58221",
          "title": "I need a formula that can extract data from Column B and Column C into Column E and Column F, provided that the correspo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58221",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58221/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58221/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58296",
          "title": "I'm trying to sum values in the 'Amount' column of a table for a given month, using a SUMIFS function in Excel. My equat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58296",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58296/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58296/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "5835",
          "title": "I have a basic table with relevant data and a requested table. How do I add the LOG VALUE from column I to the correspon",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "5835",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5835/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/5835/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58484",
          "title": "How can I count the number of times the operator, whose transfer number is 5551234, transfers a phone call to other staf",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58484",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58484/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58484/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58499",
          "title": "How can I create a formula in Excel to list all the branches in column A that do not have a corresponding tick mark in c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58499",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58499/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58499/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "585-41",
          "title": "How can I update my script so that it removes all non-numeric characters from cell values, but still preserves decimal p",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "585-41",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/585-41/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/585-41/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58523",
          "title": "How can I create an Excel formula to calculate an employee's salary paid for the days worked within each month, consider",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58523",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58523/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58523/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58568",
          "title": "I'm struggling to use the MATCH and INDEX functions correctly in Excel. I want to look up a number based on two criteria",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58568",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58568/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58568/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58570",
          "title": "How can I create a formula that counts each individual clinic visit made by a person from a large data set where each ro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58570",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58570/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58570/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58613",
          "title": "In my spreadsheet where I'm calculating ages from dates of birth, the DATEDIF formula (=DATEDIF(C2, TODAY(), \"y\")) is yi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58613",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58613/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58613/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58629",
          "title": "I need to create an Excel formula that returns the quantity of character sequences separated by commas in a cell, which ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58629",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58629/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58629/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58656",
          "title": "How can I fix a formula in Excel that is supposed to calculate lunar months by checking for two conditions: that a speci",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58656",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58656/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58656/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58667",
          "title": "I have a spreadsheet for tracking student attendance and I need the spreadsheet to automatically duplicate the 'Holiday'",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58669",
          "title": "I'm using an Index formula to retrieve data from other sheets and need to find the maximum value in an array. Additional",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58669",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58669/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58669/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58687",
          "title": "I have a horizontal table that shows the number of apples sold each month with a total of 92 apples, and a vertical tabl",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58687",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58687/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58687/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "587-38",
          "title": "How can I use VBA to fill a certain number of cells with a value based on the result from a COUNT formula? For instance,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "587-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/587-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/587-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58701",
          "title": "How can I automatically populate an office code in a specific cell on the 'Entry Tab' when a user manually enters a loca",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58701",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58701/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58701/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58723",
          "title": "How can I create a formula that examines a range in Excel to 1. Look for a particular Name and Week Start Date, 2. Ident",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58723",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58723/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58723/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58781",
          "title": "How can I sum values for each day of the week, such as all Sundays, Mondays, and so on, using my daily data from a datab",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58781",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58781/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58781/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58829",
          "title": "I need a formula to calculate car import tax based on annual CO2 emission levels and whether the fuel type is GAS or DIE",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58829",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58829/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58829/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58890",
          "title": "How can I enter numbers 1 to 5 in column F at will and retrieve corresponding data from the 'lookup' sheet into column G",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58890",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58890/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58890/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58904",
          "title": "How can I create a formula in Excel to sum the numbers in cells that contain a mix of letters and numbers, with the sum ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58904",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58904/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58904/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58942",
          "title": "I have two sets of data: Dates and Values, that are not in a sequential order. I want to create a sequential timeline in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58942",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58942/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58942/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58949",
          "title": "I need to reorganize data exported from our database where multiple rows correspond to single clients with various detai",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58949",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58949/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58949/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "58994",
          "title": "I have a column with values formatted as 'What:Who-Size-SKU' (e.g., 'FG:BJ-155-00310'), and I need to extract the 'Size'",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "58994",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58994/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/58994/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59-21",
          "title": "How can I create a macro to search for all occurrences of the word 'Total' within a worksheet and clear the contents of ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59055",
          "title": "How do I match items located in Column L with those in Column A and find the corresponding row number in Column A where ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59055",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59055/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59055/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "591-29",
          "title": "I'm attempting to write a VBA script to remove duplicate rows from a table based on a unique value in column C. Specific",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "591-29",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/591-29/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/591-29/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59129",
          "title": "How can I adjust the SUMPRODUCT formula to calculate month over month headcount in Excel, so it does not continue to cou",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59129",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59129/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59129/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59160",
          "title": "How can I compute the number of occurrences of 'Yes' or 'No' in a worksheet for a specific date using a formula, conside",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59160",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59160/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59160/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59185",
          "title": "In my Excel spreadsheet, I have alternating rows of names and numbers. I've been using an HLOOKUP formula to search for ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59185",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59185/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59185/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59196",
          "title": "I need a formula to determine which column contains the highest value in a row, and then return the heading of that colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59196",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59196/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59196/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "592-34",
          "title": "How can I use a VB (Visual Basic) script in Excel to split the contents of a single column into multiple columns whereve",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "592-34",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/592-34/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/592-34/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59224",
          "title": "I need assistance with an Excel spreadsheet where I have a range of predetermined dates in columns C and D. In column E,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59224",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59224/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59224/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59259",
          "title": "How can I create an interactive Excel formula that will calculate the sum of year-to-date amounts based on a selected mo",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59259",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59259/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59259/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59279",
          "title": "How can I create a formula in Excel to extract a date of birth from a string, where the first six characters represent t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59279",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59279/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59279/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59340",
          "title": "How can I modify my current SUMPRODUCT formula in Excel, which is =SUMPRODUCT((B2:B20=17)*(D2:K20)), so that it only sum",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59340",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59340/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59340/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59358",
          "title": "How can I search through two ranges in Excel (B2:B23 and D2:D23), identify the first occurrence where a value in range B",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59358",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59358/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59358/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59433",
          "title": "How do I create a formula to extract a specific site code from within varying lengths of text strings? For instance, fro",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59433",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59433/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59433/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59511",
          "title": "I need to know the formula for incrementing a value by 1 for every 1,000, without using the XLOOKUP formula. This is bec",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59511",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59511/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59511/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59595",
          "title": "In my Excel spreadsheet, I have a column A with non-continuous but sequentially increasing dates and a column B with poi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59595",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59595/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59595/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "596-20",
          "title": "How can I use VBA to identify numbers that are nearly identical in the same row across a range of columns from N to DS i",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "596-20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/596-20/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/596-20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "596-28",
          "title": "How can I create a macro in Excel to merge a cell in column B that begins with the text 'lo.:' with the cell above it, b",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "596-28",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/596-28/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/596-28/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59626",
          "title": "In cell D17, I have a price calculation for specific dates. The discounts can be applied in two ways and are input in se",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59626",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59626/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59626/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59639",
          "title": "I have data within a single cell, formatted as a continuous string of characters (e.g., 'abcdefgh'), and I need to separ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59639",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59639/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59639/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59643",
          "title": "How do I calculate the average duration of a transaction for each hour of each day directly from two columns in Excel, o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59643",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59643/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59643/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59654",
          "title": "How can I calculate the patients who returned to the emergency room within 24 hours after their previous discharge using",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59654",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59654/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59654/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59667",
          "title": "How do I create a formula in Excel that will populate column D based on conditions related to the data in columns B and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59667",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59667/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59667/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59734",
          "title": "I have two datasets, one with client data and another with episode data, where each client can access up to four program",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59734",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59734/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59734/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59743",
          "title": "I need to modify the formula in column G of my Excel spreadsheet so that it calculates the difference between F and J if",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59743",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59743/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59743/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59787",
          "title": "I need to display the number of days an action item is past its due date in a spreadsheet I'm working on. The due date o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59787",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59787/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59787/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59791",
          "title": "How can I modify the formula in column D of my Excel workbook to remove decimal points from the numbers without rounding",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59791",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59791/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59791/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59794",
          "title": "How can I correctly split the information from one column (column A labeled 'INITIAL INFORMATION Column1') into multiple",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59794",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59794/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59794/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "598-14",
          "title": "I am trying to find a Visual Basic for Applications (VBA) solution to determine the first and last row numbers where val",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "598-14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/598-14/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/598-14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "598-46",
          "title": "I need a VBA macro that can be used in an Excel workbook to delete any row where the data in column B does not have any ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "598-46",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/598-46/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/598-46/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59884",
          "title": "How can I consolidate repeating first values in a list, count them, and then transpose the corresponding second values f",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59884",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59884/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59884/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "599-9",
          "title": "In my Excel workbook, column G contains a list of dates that changes with every report run. I need a script that would t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "599-9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/599-9/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/599-9/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59902",
          "title": "How can I adjust my Index Match formula to calculate the days since our staff last made a sale, with the dates sorted in",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59902",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59902/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59902/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59932",
          "title": "I need help to perform a lookup in Excel that finds the closest row with a date that is before the specified lookup date",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59932",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59932/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59932/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59969",
          "title": "How can I sum monetary data in Excel by matching a specific text in the header row and a term in column M, and then get ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59969",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59969/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59969/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "59979",
          "title": "How can I create a formula that assigns specific employees from my call data a code of 'Sup' if they are on a given list",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "59979",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59979/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/59979/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "60-7",
          "title": "How can I update the data in a 'Consolidated Tracker' tab in Excel by combining the rows from an 'Existing Task' sheet a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "60-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/60-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/60-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "601-4",
          "title": "How can I create a macro in Excel to copy data from a sheet named 'GL Numbers' to a sheet named 'extract items NA' if tw",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "601-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/601-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/601-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "61-1",
          "title": "How can I create a VBA macro that will append '-001' to the end of each item name in column D, as this will be included ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "61-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/61-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/61-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "61-4",
          "title": "How can I identify consecutive negative values in a column for specific stock names, and then merge the related data suc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "61-4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/61-4/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/61-4/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "62-3",
          "title": "How do I consolidate the amounts for columns C and D by their respective dates, with the result showing the merged amoun",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "62-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/62-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/62-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "62-48",
          "title": "I am working on an Excel list that spans over 9000 rows with several data points laid out horizontally, which include du",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "62-48",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/62-48/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/62-48/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "6239",
          "title": "I'm creating an incentive program to reward my employees based on their performance and need help with setting it up in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "6239",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6239/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6239/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "6435",
          "title": "How can I automatically calculate the number of days a product remains in working condition after it is repaired, taking",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "6435",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6435/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6435/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "66-1",
          "title": "In my stock list, products have their dimensions mentioned at the end of their description in the note. I want to calcul",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "66-1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/66-1/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/66-1/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "66-24",
          "title": "How do I write a macro to select and extract data from the 'Imported Data' sheet to the 'Items Older than 30 days' sheet",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "66-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/66-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/66-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "6698",
          "title": "How do I modify my Excel formula in cell G4 that currently counts the number of occurrences where either the value in co",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "6698",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6698/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/6698/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "68-47",
          "title": "How can I apply an AutoFilter in Excel to sort based on cell color but exclude the first row and the last three rows? An",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "68-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/68-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/68-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "69-17",
          "title": "I'm looking to convert a COUNTIFS function that I currently use in column AW to a macro, because running this function o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "69-17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/69-17/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/69-17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "7-5",
          "title": "How can I count the number of green and red cells within a specific row (D4:AO4) of an Excel 2010 worksheet, where there",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "7-5",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7-5/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7-5/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "70-3",
          "title": "How can I perform subtraction only on cells that are highlighted, and is it possible to make a flexible VBA formula that",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "70-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/70-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/70-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "71-30",
          "title": "How can I write VBA code to transpose each product number from a horizontal to a vertical layout and align all correspon",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "71-30",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/71-30/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/71-30/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "73-45",
          "title": "I am looking for a macro that can identify and highlight specific text, numbers, or parts of text and numbers in column ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "73-45",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/73-45/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/73-45/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "74-47",
          "title": "I need assistance with a VBA code that will append a comma at the end of every numeric value within my Excel workbook. F",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "74-47",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/74-47/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/74-47/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "76-42",
          "title": "I need to calculate the time difference in column D based on start times in column B and end times in column C using Exc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "76-42",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/76-42/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/76-42/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "76-43",
          "title": "I'm seeking a VBA code solution to find and replace a specific word within a single cell (A1&A2) in Excel. In this cell,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "76-43",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/76-43/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/76-43/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "7665",
          "title": "How can I transform data from vertical format in column H to a horizontal format in cells Q2:V2, making sure to only ext",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "7665",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7665/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7665/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "79-7",
          "title": "I am struggling to use VBA to find and replace all instances of double quotes with nothing across all columns in a works",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "79-7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/79-7/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/79-7/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "7902",
          "title": "I need assistance with constructing a formula in my worksheet. I want to compare two values in a row, and based on that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "7902",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7902/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/7902/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "80-42",
          "title": "How can I consolidate data from over 20 sheets with substantial data into one sheet, including only specific columns, an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "80-42",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/80-42/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/80-42/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "81-41",
          "title": "I need a VBA macro for my Excel sheet that will remove all instances of a single standalone dot (.) and the specific tex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "81-41",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/81-41/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/81-41/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "82-30",
          "title": "How can I modify my VBA code to only copy and paste whole numbers from a range of columns (A to F) in the 'Raw Data' she",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "82-30",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/82-30/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/82-30/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "82-38",
          "title": "On my SH1 sheet, there are multiple headers labeled 'INVOICE NO' with a number underneath each for different ranges. In ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "82-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/82-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/82-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "84-38",
          "title": "In my 'Compare' sheet, I have data in columns A and B which I am trying to sort independently in ascending order using V",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "84-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/84-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/84-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "84-40",
          "title": "I need to sum the amounts in the entire columns C and D for each sheet, excluding column B, in an Excel workbook. These ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "84-40",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/84-40/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/84-40/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "85-38",
          "title": "How can I automate the process of deleting rows in an Excel table if the person is listed as a 'Child living at home' an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "85-38",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/85-38/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/85-38/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "8942",
          "title": "In my Excel workbook, I have an 'overview' sheet with an Index match set up in cells B2 and C2, and a data validation li",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "8942",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/8942/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/8942/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "91-3",
          "title": "How can I transpose multiple sheets in an Excel file where each contains multiple words in column B, and create a final ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "91-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/91-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/91-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "91-34",
          "title": "How can I create a macro in Excel that identifies and deletes only one entry of a person (same last name, first name, an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "91-34",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/91-34/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/91-34/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "9111",
          "title": "I have two lists with names and corresponding IDs. How can I fill a table where I need to return specific IDs from these",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "9111",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9111/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9111/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "9391",
          "title": "Is it possible to create an Excel formula that takes a date from cell B1 on a 'Front' sheet, searches for that date in c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "9391",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9391/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9391/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "94-28",
          "title": "I have written a VBA macro that is supposed to remove spaces and round numbers to remove decimals in column C of a works",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "94-28",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/94-28/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/94-28/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "94-8",
          "title": "How can I use VBA to eliminate rows in an Excel sheet when the Item column doesn't match any item in a pre-defined list ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "94-8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/94-8/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/94-8/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "9448",
          "title": "How can I create a formula in column U of my stock control spreadsheet that will retrieve and display the header value o",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "9448",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9448/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9448/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "9569",
          "title": "I need a formula to identify the earliest non-zero value entered for each row, specifically within column 'AO,' from a r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "9569",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9569/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9569/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "97-36",
          "title": "How can I create a VBA code that deletes all rows in an Excel sheet above a specified substring, starting from the first",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "97-36",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/97-36/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/97-36/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "9726",
          "title": "How can I populate the 'Visit date' column in Table 2 with the date from the 'Visit date' column in Table 1 that is imme",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "9726",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9726/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/9726/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "98-21",
          "title": "I need to distribute an amount specified in cell M2 across several items within my Excel sheet, specifically allocating ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "98-21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/98-21/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/98-21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "99-24",
          "title": "How can I filter a large dataset in Excel to display records within the 2019 to 2023 date range where the 'Date Paid' fi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "99-24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/99-24/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/99-24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "99-3",
          "title": "How can I modify my VBA code, which currently deletes rows containing the text 'Matched' across certain cells in a range",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "99-3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/99-3/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/99-3/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_10214",
          "title": "How do I set up conditional formatting in Excel so that the number in column C turns red if it is lower or higher than t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_10214",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_10214/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_10214/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_11072",
          "title": "I'd like to know how to use conditional formatting between 2 cells depending on what the value is of the 2 cells and wha",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_11072",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_11072/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_11072/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_11177",
          "title": "How can I highlight cells with color based on different times?  I have provided more details in the attached sheet. Plea",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_11177",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_11177/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_11177/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_12196",
          "title": "How can I create a conditional format to highlight cells in green that represent orders over 75% of the weekly delivery ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_12196",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12196/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12196/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_12429",
          "title": "How can I apply conditional formatting to a cell in the 'score 1' or 'score 2' columns in my Excel 2016 spreadsheet, so ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_12429",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12429/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12429/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_12518",
          "title": "How do I set up conditional formatting to highlight cells with a value less than 1 in red, but do not highlight blank ce",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_12518",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12518/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_12518/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_13024",
          "title": "How can I automatically highlight each staff's Saturday or Sunday off in an attendance/holiday planner, using Conditiona",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_13024",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13024/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13024/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_13984",
          "title": "How can I highlight the Issue Certificate column (E) in red when all three tests (theory, practical, and oral) have been",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_13984",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13984/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13984/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_13993",
          "title": "Is there a way to apply conditional formatting to a specific column based on a specific header? Query: Need to put condi",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_13993",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13993/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_13993/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_15439",
          "title": "How can I create a conditional formatting rule that changes the cell color to yellow in columns E to J when column D has",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_15439",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_15439/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_15439/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_15571",
          "title": "How can I highlight duplicate SKUs in column I with red color without highlighting the same column? I have a list of SKU",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_15571",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_15571/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_15571/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_16840",
          "title": "How can I use conditional formatting to highlight cells in column C with yellow that are not in the 'Complete' or 'Refer",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_16840",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_16840/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_16840/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_17940",
          "title": "How can I apply conditional formatting to a range of cells (F3:K22) to represent progress on various lists of activities",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_17940",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_17940/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_17940/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_18814",
          "title": "How can I apply conditional formatting to highlight Sundays in my attendance sheet with yellow, and have it automaticall",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_18814",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_18814/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_18814/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_18989",
          "title": "How can I create a conditional formatting rule that applies a yellow/white background color for alternating years based ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_18989",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_18989/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_18989/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_19190",
          "title": "How can I highlight all the cells in a column based on the weekday name that is highlighted in green? For example, if it",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_19190",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_19190/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_19190/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_21040",
          "title": "How can I highlight countrywise holiday details in a sheet with yellow based on a holiday list table using conditional f",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_21040",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_21040/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_21040/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_21524",
          "title": "How do I find the differences between two ID columns in Excel and pull the corresponding names? I have a new ID column (",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_21524",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_21524/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_21524/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_22493",
          "title": "How can I change the cell color in column R to red when the cell's value is above zero? I have provided an example in th",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_22493",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_22493/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_22493/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_24632",
          "title": "How can I automatically highlight cells in Excel with yellow based on start and end dates listed in separate columns, to",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_24632",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_24632/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_24632/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_24784",
          "title": "I have a spreadsheet where in Row 3 the correct results are displayed, and other rows consist of people's choice selecti",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_24784",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_24784/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_24784/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_25745",
          "title": "How can I use conditional formatting to change the fill color of a cell based on the values of two other cells? I have t",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_25745",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_25745/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_25745/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_26100",
          "title": "How can I change the cell color in D3 based on the value in C3, and change the cell color in D10 based on the date in C1",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_26100",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_26100/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_26100/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_26944",
          "title": "How do I set up a conditional format that changes a cell's color to red if the number of days in a cell formatted as 'X ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_26944",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_26944/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_26944/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_28312",
          "title": "In my Excel spreadsheet, I have a column A with tasks listed and other columns represent dates on which I have worked on",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_28312",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_28312/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_28312/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_28766",
          "title": "I need help creating a formula in Excel that will allow me to identify and highlight with yellow only the second and sub",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_28766",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_28766/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_28766/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_2919",
          "title": "I am using conditional formatting to highlight the range of the scores but would like a way to quickly identify the best",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_2919",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_2919/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_2919/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_29431",
          "title": "How can I apply conditional formatting to cells B3:B43 in Excel such that if cell L3 is greater than 10,00,000, then B3:",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_29431",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_29431/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_29431/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_357",
          "title": "How can I apply Conditional Formatting to a cell based on two criteria, one for a row and one for a column, in Excel? Ho",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_357",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_357/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_357/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_3575",
          "title": "How can I color a cell red to a specific deliverable's deadline? The deadline week is in Column D and the week numbers a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_3575",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_3575/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_3575/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_3712",
          "title": "How can I apply conditional formatting to a cell based on the number of days left to pay an invoice, where the cell disp",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_3712",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_3712/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_3712/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_4909",
          "title": "Currently, the conditional formatting works by checking the value in the cells of row 29 and highlighting that value in ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_4909",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_4909/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_4909/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_513",
          "title": "I am trying to prepare a bill calendar that has bills on their due dates. I want to set up conditional formatting so tha",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_513",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_513/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_513/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_5348",
          "title": "How can I automate my attendance list in Excel where the SELECT LIST column is dynamic and the ATTENDANCE LIST column is",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_5348",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_5348/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_5348/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_6540",
          "title": "I have created a spreadsheet for a comparison of rates between different supplier. I have applied conditional formatting",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_6540",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_6540/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_6540/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_7349",
          "title": "How can I apply conditional formatting to highlight cells in two categories (Cat1 App & Cat2 Claims) based on specific c",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_7349",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_7349/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_7349/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_7938",
          "title": "How can I use conditional formatting to highlight cells D8, G8, and H8 in red to identify non-working days or faulty ent",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_7938",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_7938/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_7938/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_8276",
          "title": "How can I set up a conditional format to color a cell yellow when it contains a hyperlink and leave it with no color if ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_8276",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_8276/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_8276/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_8830",
          "title": "I am needing to highlight cells in red where fuel types don't match up. In the Excel file, I have two columns, A and B w",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_8830",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_8830/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_8830/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v1-full-912",
          "taskId": "CF_9945",
          "title": "I need to apply conditional formatting to cell C3 so that it turns green if a range of cells (C4-C13) meets the followin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v1",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v1",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v1-912",
              "SPREADSHEETBENCH_TASK_ID": "CF_9945",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v1-912-stage.json",
            "docs/eval/spreadsheetbench-v1-912-copy-input-baseline.json",
            "docs/eval/official-benchmark-readiness.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_9945/agent/task.json",
            ".tmp/official-benchmarks/staged-v1-912/tasks/CF_9945/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        }
      ]
    },
    {
      "id": "spreadsheetbench-v2-full-321",
      "title": "SpreadsheetBench 2 full workflow benchmark",
      "taskCount": 321,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 321,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: deleted rows causing #REF! errors,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/01_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/01_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_01_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: deleted rows or cells causing #REF",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/02_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/02_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_02_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: deleted rows causing #REF! errors,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/03_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/03_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_03_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: deleted rows causing #REF! errors,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/04_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/04_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_04_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: deleted rows causing #REF! errors,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/05_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/05_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_05_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/06_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/06_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_06_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/07_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/07_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_07_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/08_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/08_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_08_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/09_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/09_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_09_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_01",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_02",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_03",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_04",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_05",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_06",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_07",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_08",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_09",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Debugging/10_10",
          "title": "Please audit and fix this file thoroughly. Errors may include but are not limited to: formula logic errors or inconsiste",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Debugging/10_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Debugging_10_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/01_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/01_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/01_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/01_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/01_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/01_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/01_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/01_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/01_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/01_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_01_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/02_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/02_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/02_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/02_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/02_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/02_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/02_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/02_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/02_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/02_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_02_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/03_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/03_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/03_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/03_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/03_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/03_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/03_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/03_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/03_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/03_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_03_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/04_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/04_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/04_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/04_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/04_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/04_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/04_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/04_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/04_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/04_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_04_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/05_01",
          "title": "Complete the AIF financial model based on the provided assumptions. Ensure the existing structure, layout, and formattin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/05_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/05_02",
          "title": "Complete the AIF financial model based on the provided assumptions. Ensure the existing structure, layout, and formattin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/05_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/05_03",
          "title": "Complete the AIF financial model based on the provided assumptions. Ensure the existing structure, layout, and formattin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/05_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/05_04",
          "title": "Complete the AIF financial model based on the provided assumptions. Ensure the existing structure, layout, and formattin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/05_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/05_05",
          "title": "Complete the AIF financial model based on the provided assumptions. Ensure the existing structure, layout, and formattin",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/05_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_05_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/06_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/06_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/06_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/06_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/06_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/06_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/06_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/06_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/06_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/06_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_06_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/07_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/07_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/07_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/07_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/07_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/07_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/07_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/07_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/07_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/07_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_07_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/08_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/08_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/08_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/08_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/08_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/08_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/08_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/08_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/08_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/08_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_08_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/09_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/09_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/09_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/09_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/09_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/09_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/09_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/09_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/09_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/09_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_09_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/10_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/10_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/10_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/10_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/10_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/10_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/10_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/10_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/10_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/10_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_10_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/11_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/11_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/11_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/11_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/11_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/11_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/11_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/11_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/11_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/11_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_11_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/12_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/12_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/12_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/12_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/12_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/12_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/12_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/12_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/12_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/12_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_12_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/13_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/13_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/13_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/13_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/13_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/13_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/13_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/13_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/13_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/13_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_13_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/14_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/14_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/14_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/14_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/14_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/14_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/14_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/14_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/14_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/14_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_14_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/15_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/15_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/15_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/15_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/15_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/15_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/15_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/15_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/15_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/15_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_15_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/16_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/16_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/16_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/16_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/16_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/16_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/16_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/16_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/16_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/16_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_16_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/17_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/17_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/17_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/17_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/17_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/17_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/17_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/17_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/17_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/17_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_17_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/18_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/18_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/18_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/18_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/18_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/18_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/18_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/18_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/18_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/18_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_18_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/19_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/19_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/19_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/19_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/19_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/19_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/19_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/19_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/19_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/19_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_19_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/20_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/20_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/20_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/20_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/20_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/20_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/20_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/20_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Financial_Model/20_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure the existing structure, layout, and formatting of",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Financial_Model/20_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Financial_Model_20_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_01",
          "title": "Complete the bond accounting schedule for Coastal Manufacturing Corp's 6-year, 4% coupon discount bond.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_02",
          "title": "Complete the OID bond amortization analysis for XYZ Corporation. Calculate amortization, IS and BS results for GAAP and ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_03",
          "title": "Complete the zero-coupon bond accounting analysis by filling in all empty cells in the financial statements and balance ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_04",
          "title": "Complete the OID bond amortization and deferred tax analysis for TechVenture Inc.'s 7-year zero-coupon bond.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_05",
          "title": "Complete the bond accounting model for ABC Corp's 4-year bond issuance. Fill in all empty cells.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_06",
          "title": "Complete the bond accounting analysis. Fill in all empty cells. Use the effective interest method. For tax purposes, the",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_07",
          "title": "Complete the bond accounting analysis using the effective interest method. Fill in all empty cells.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_08",
          "title": "Complete the bond accounting model. Calculate the implied discount rate (IRR of provided cash flows), amortization sched",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/01_09",
          "title": "Complete the bond accounting model for a 4-year premium bond with early extinguishment in year 4.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/01_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_01_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_04",
          "title": "Use the average balance method for interest expense.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_05",
          "title": "Complete the debt waterfall model. The company retains exactly its operating cash requirement as the ending cash balance",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/02_06",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/02_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_02_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/03_01",
          "title": "Complete the M&A consolidation analysis by filling all blank cells in the financial statements.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/03_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/03_02",
          "title": "Complete the M&A consolidation exercise. Fill in all empty cells. Assume US GAAP (ASC 805); measure NCI at fair value.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/03_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/03_03",
          "title": "Complete the M&A consolidation model. Calculate acquisition-date balance sheet (3/31/2019) and FY 2020 consolidated fina",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/03_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/03_04",
          "title": "Complete the M&A consolidation analysis by filling in the consolidation worksheet. Calculate the fair value adjustments,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/03_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_03_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/04_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/04_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_04_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_04_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/04_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/04_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_04_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_04_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/05_01",
          "title": "Complete the deferred tax analysis by filling all empty cells in the Income Statement, Balance Sheet, Taxable Income Cal",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/05_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_05_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_05_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/05_02",
          "title": "Complete the deferred tax analysis. Fill in all empty cells for Years 2-4. Leave Tax Accounting income statement rows an",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/05_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_05_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_05_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_06",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_07",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_08",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_09",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_11",
          "title": "Forecast Cash as prior period Cash plus Changes in Working Capital.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_11",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_11/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_11/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_12",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_12/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_13",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_13",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_13/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_13/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_14",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_14",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_14/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_14/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_15",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_15",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_15/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_15/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_16",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_16/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_16/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_17",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_17",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_17/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_17/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_18",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_18/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_18/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_19",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_19",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_19/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_19/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_20",
          "title": "Complete the revenue forecast model. Project each segment using its historical CAGR from Year 1 to Year 3. Fill in all e",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_20",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_20/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_20/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_21",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_21",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_21/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_21/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_22",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_22",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_22/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_22/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_23",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_23",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_23/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_23/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_24",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_24/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_24/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/06_25",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/06_25",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_25/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_06_25/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/07_01",
          "title": "Complete the equity method accounting model. Fill in all empty cells. Account for deferred tax effects from the dividend",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/07_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/07_02",
          "title": "Complete the equity method investment analysis by calculating the purchase price allocation, balance sheet impacts, inco",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/07_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/07_03",
          "title": "Complete the equity method accounting exercise. Acquisition closes 3/31/2016 (start of Q1). The Q1 End balance sheet ref",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/07_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_07_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/08_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/08_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/08_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/08_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/08_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/08_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_08_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/09_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/09_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/09_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/09_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/09_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/09_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/09_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/09_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_09_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/10_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/10_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_10_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_10_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/10_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/10_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_10_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_10_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/11_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/11_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/11_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/11_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/11_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/11_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/11_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/11_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_11_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/12_01",
          "title": "Complete the NOL carryforward calculations for both scenarios. Fill in all empty cells. Key assumptions (tax rate, NOL u",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/12_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_12_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_12_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_01",
          "title": "Normalize the operating performance data by removing nonrecurring items from the income statement, complete the financia",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_02",
          "title": "Normalize the historical financial statements by removing non-recurring items specified in the adjustments section. Calc",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_03",
          "title": "Complete the normalized income statement and key metrics for all quarters (Q1 2024 through Q2 2025). Adjust historical v",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_04",
          "title": "Normalize earnings for non-recurring items, calculate financial metrics, and project 2023-2025 using the specified avera",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_05",
          "title": "Normalize the financial data for non-recurring items, calculate metrics, and project future quarters using the specified",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_06",
          "title": "Normalize the financial statements for non-recurring items, calculate historical metrics, and forecast 2023E-2025E using",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_07",
          "title": "Normalize the 2020-2021 financial data, calculate historical metrics, and project 2022E-2025E using historical averages.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/13_08",
          "title": "Normalize the financial statements, calculate operating metrics, and forecast 2022-2023 per specified averaging methods.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/13_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_13_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_06",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_07",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_08",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/14_09",
          "title": "Concessions are 0% in Years 3–5.",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/14_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_14_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/15_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/15_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/15_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/15_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/15_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/15_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/15_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/15_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_15_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_01",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_01",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_01/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_01/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_02",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_02",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_02/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_02/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_03",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_03",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_03/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_03/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_04",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_04",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_04/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_04/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_05",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_05",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_05/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_05/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_06",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_06",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_06/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_06/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_07",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_07",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_07/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_07/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_08",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_08",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_08/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_08/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_09",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_09",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_09/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_09/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_10",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_10",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_10/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_10/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_11",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_11",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_11/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_11/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Template/16_12",
          "title": "Complete the financial model based on the provided assumptions. Ensure that all formulas, references, and calculations a",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Template/16_12",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_12/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Template_16_12/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 126",
          "title": "Create 2 Earnings Per Share charts using the provided dataset in the Surprise sheet (rows 50-71). The first chart is a m",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 126",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_126/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_126/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 127",
          "title": "Create a stacked column chart showing the Analyst Target Price of AppLovin where each column shows the distribution of '",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 127",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_127/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_127/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 129",
          "title": "Create a football field valuation chart titled 'Project Mars - Standalone Valuation Summary'. Display horizontal bars sh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 129",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_129/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_129/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 131",
          "title": "Create a waterfall (bridge) chart titled 'EBITDA Waterfall ($M)' showing the movement in EBITDA from FY22 to FY23. Start",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 131",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_131/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_131/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 132",
          "title": "Create four multi-series line charts from the data in Sheet1. Each chart should have the four cases (Street, Mgmt-Base, ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 132",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_132/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_132/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 133",
          "title": "Create a speedometer/gauge chart (half-doughnut style) for the Battery Charge Progress (%) value in Sheet1. The chart sh",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 133",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_133/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_133/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 134",
          "title": "Using the ellipse data in columns A and B, create a scatter chart with blue markers forming the ellipse. Label the six v",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 134",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_134/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_134/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1411527",
          "title": "Create a Line chart in the Count sheet titled \"Invoice Count Burndown August 2023\". The chart should display daily data ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1411527",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1411527/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1411527/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1417365",
          "title": "Create a stacked column chart to compare sales for Type A and Type B across Q1–Q4 for the years 2022 and 2023.\nThe colum",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1417365",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1417365/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1417365/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1419935",
          "title": "Using the volleyball game raw data (with rows: Points, Team (where Team is \"ETV\" or \"Opp\"), Event, Player, and Position,",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1419935",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1419935/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1419935/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1420776",
          "title": "Consolidate the data from February 2024 and March 2024 to create a multi series line chart to show the measurement value",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1420776",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1420776/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1420776/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1421372",
          "title": "Create a dashboard on a sheet named “Summary” to consolidate data from 2-hour racing events, organizing the data into 5-",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1421372",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1421372/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1421372/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1422635",
          "title": "Create a scatter chart from my blood sugar data using the Date and Value columns. Add helper columns with formulas that ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1422635",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1422635/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1422635/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1423401",
          "title": "Create five separate clustered column charts displaying two data series side-by-side: ‘Foo’ and ‘Faa’.\nEach chart should",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1423401",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1423401/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1423401/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1426290",
          "title": "Based on the daily patrol times, calculate the average patrol duration for each day and place it in a new \"Average\" colu",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1426290",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1426290/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1426290/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1428419",
          "title": "1. Create a chart titled \"Chart A Scores\". Each row of data from the Chart A column should appear as a single square mar",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1428419",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1428419/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1428419/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1430721",
          "title": "Create a bubble chart titled 'Proportional value of Transaction' that bases the size of the bubbles on the Proportional ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1430721",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1430721/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1430721/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1431124",
          "title": "Create a dynamic summary section in the “Forecasting Summary Data” sheet that first includes a drop-down list for select",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1431124",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1431124/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1431124/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1433875",
          "title": "Create a line chart for the monthly orders data grouped per year with a data table below inside the chart.\nExclude the \"",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1433875",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1433875/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1433875/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1434951",
          "title": "Create a line chart that compares the sales amounts of two series across years (primary x-axis) and months (in text form",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1434951",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1434951/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1434951/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 1437004",
          "title": "Pull the data from the \"Sharepoint Tracker\" sheet, count the total number of TRUE values in each category for each corre",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 1437004",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1437004/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_1437004/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 90",
          "title": "Create a Pareto chart titled 'Defect Analysis'. Use the defect categories and counts from the 'Data' sheet. Show counts ",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 90",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_90/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_90/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 95",
          "title": "Create a bubble chart titled 'Product Portfolio Analysis' based on the data available on the sheet. Plot products with r",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 95",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_95/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_95/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        },
        {
          "familyId": "spreadsheetbench-v2-full-321",
          "taskId": "Visualization/Task 96",
          "title": "Create a sunburst chart titled 'Expense Breakdown'. Show a 3-level hierarchy of expenses: department → category → subcat",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:spreadsheetbench-v2",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "SPREADSHEETBENCH_TRACK": "spreadsheetbench-v2",
              "SPREADSHEETBENCH_STAGE_ROOT": ".tmp/official-benchmarks/staged-v2-full",
              "SPREADSHEETBENCH_TASK_ID": "Visualization/Task 96",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "docs/eval/spreadsheetbench-v2-full-ingest.json",
            "docs/eval/spreadsheetbench-v2-full-stage.json",
            "docs/eval/spreadsheetbench-v2-stage-smoke.json",
            "docs/eval/spreadsheetbench-v2-run-smoke.json",
            "docs/eval/spreadsheetbench-chart-visual-probe.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_96/agent/task.json",
            ".tmp/official-benchmarks/staged-v2-full/tasks/Visualization_Task_96/evaluator/evaluator.json"
          ],
          "blockers": [
            "SpreadsheetBench task is ready for the generic prod browser workbook adapter but lacks a passing prod receipt."
          ]
        }
      ]
    },
    {
      "id": "bankertoolbench-full-100",
      "title": "BankerToolBench full investment-banking benchmark",
      "taskCount": 100,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 100,
      "runnableProdBrowserTasks": 100,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "707cba99-59a7-47bd-bc4d-7f36212e99f3",
          "title": "I am an investment banking analyst in the REGAL coverage group at XYZ bank. I'm covering Real Estate, Gaming and Lodging",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "707cba99-59a7-47bd-bc4d-7f36212e99f3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/707cba99-59a7-47bd-bc4d-7f36212e99f3.json",
            "docs/eval/fresh-room/FR-020/tasks/707cba99-59a7-47bd-bc4d-7f36212e99f3/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "4e23ce4e-be6f-4d56-ab7b-2878a801a648",
          "title": "I am an investment banking analyst in the REGAL coverage group at XYZ bank. I'm covering Real Estate, Gaming and Lodging",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "4e23ce4e-be6f-4d56-ab7b-2878a801a648",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/4e23ce4e-be6f-4d56-ab7b-2878a801a648.json",
            "docs/eval/fresh-room/FR-020/tasks/4e23ce4e-be6f-4d56-ab7b-2878a801a648/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "11e08646-55be-41cd-a0b8-24d1529b76aa",
          "title": "Hey, just got pulled out of a call with Disney's corp dev team. They're seriously looking at Warner Bros. Discovery as a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "11e08646-55be-41cd-a0b8-24d1529b76aa",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/11e08646-55be-41cd-a0b8-24d1529b76aa.json",
            "docs/eval/fresh-room/FR-020/tasks/11e08646-55be-41cd-a0b8-24d1529b76aa/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "7f5b9960-5931-4fb7-ac69-a5d0a954b1cc",
          "title": "My VP just pulled me into his office - we're pitching FedEx to a potential strategic acquirer next week and he needs a s",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "7f5b9960-5931-4fb7-ac69-a5d0a954b1cc",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/7f5b9960-5931-4fb7-ac69-a5d0a954b1cc.json",
            "docs/eval/fresh-room/FR-020/tasks/7f5b9960-5931-4fb7-ac69-a5d0a954b1cc/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "9781216f-b98d-4537-9ab8-f0e62248a6ab",
          "title": "Today is December 31, 2024. You are a 2nd year analyst at an investment bank and are asked to perform an LBO valuation o",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "9781216f-b98d-4537-9ab8-f0e62248a6ab",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/9781216f-b98d-4537-9ab8-f0e62248a6ab.json",
            "docs/eval/fresh-room/FR-020/tasks/9781216f-b98d-4537-9ab8-f0e62248a6ab/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c0bf2e84-2400-47bf-9dd2-aacfb53a2090",
          "title": "Activist investors in NYT would like to exit their position, and are pushing the company to make a sale. You are an inve",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c0bf2e84-2400-47bf-9dd2-aacfb53a2090",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c0bf2e84-2400-47bf-9dd2-aacfb53a2090.json",
            "docs/eval/fresh-room/FR-020/tasks/c0bf2e84-2400-47bf-9dd2-aacfb53a2090/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "129ab204-5e9f-414d-9951-94a11eda8785",
          "title": "Your MD informed you that a buyer has reached out and is exploring a potential acquisition of Meta, Inc. The buyer is lo",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "129ab204-5e9f-414d-9951-94a11eda8785",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/129ab204-5e9f-414d-9951-94a11eda8785.json",
            "docs/eval/fresh-room/FR-020/tasks/129ab204-5e9f-414d-9951-94a11eda8785/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "b957a435-13bf-469b-b4de-fa55e3b15edd",
          "title": "You are an Analyst at an investment banking firm. Your VP has notified you of an upcoming meeting with Okta for a valuat",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "b957a435-13bf-469b-b4de-fa55e3b15edd",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/b957a435-13bf-469b-b4de-fa55e3b15edd.json",
            "docs/eval/fresh-room/FR-020/tasks/b957a435-13bf-469b-b4de-fa55e3b15edd/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "990cf5d7-0db5-456e-9766-c9d8a6640e3a",
          "title": "You are an Associate at a bulge bracket investment banking firm. Using the provided historical financials and estimates ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "990cf5d7-0db5-456e-9766-c9d8a6640e3a",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/990cf5d7-0db5-456e-9766-c9d8a6640e3a.json",
            "docs/eval/fresh-room/FR-020/tasks/990cf5d7-0db5-456e-9766-c9d8a6640e3a/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "85670d9d-a809-4f32-a3ff-d93845cc17ea",
          "title": "You are a healthcare investment banking analyst at xyz bank and you have been tasked with creating an outside in model f",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "85670d9d-a809-4f32-a3ff-d93845cc17ea",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/85670d9d-a809-4f32-a3ff-d93845cc17ea.json",
            "docs/eval/fresh-room/FR-020/tasks/85670d9d-a809-4f32-a3ff-d93845cc17ea/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "b8be8b2f-e6c2-4137-98e8-a09a3a9860f7",
          "title": "You are a healthcare investment banking analyst at xyz bank and you have been tasked with creating an outside in model f",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "b8be8b2f-e6c2-4137-98e8-a09a3a9860f7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/b8be8b2f-e6c2-4137-98e8-a09a3a9860f7.json",
            "docs/eval/fresh-room/FR-020/tasks/b8be8b2f-e6c2-4137-98e8-a09a3a9860f7/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "4a3bf36b-caeb-4825-a60c-4c416b4d555c",
          "title": "You are tasked with building a LBO model for Heidrick & Struggles (NASDAQ: HSII), strictly utilizing the attached materi",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "4a3bf36b-caeb-4825-a60c-4c416b4d555c",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/4a3bf36b-caeb-4825-a60c-4c416b4d555c.json",
            "docs/eval/fresh-room/FR-020/tasks/4a3bf36b-caeb-4825-a60c-4c416b4d555c/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "a31173e3-e8aa-4ddb-b0d9-e4e7055c950b",
          "title": "You are an Equity Research associate and want to Build a DCF Sum-of-the-Parts (SOTP) valuation for Alphabet Inc. (Google",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "a31173e3-e8aa-4ddb-b0d9-e4e7055c950b",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/a31173e3-e8aa-4ddb-b0d9-e4e7055c950b.json",
            "docs/eval/fresh-room/FR-020/tasks/a31173e3-e8aa-4ddb-b0d9-e4e7055c950b/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "eefb52c1-ee12-4d15-9fbe-7f894705d503",
          "title": "You're a junior investment banker in the consumer / retail group at a bulge bracket firm. One of the strategic / corpora",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "eefb52c1-ee12-4d15-9fbe-7f894705d503",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/eefb52c1-ee12-4d15-9fbe-7f894705d503.json",
            "docs/eval/fresh-room/FR-020/tasks/eefb52c1-ee12-4d15-9fbe-7f894705d503/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "06c284ef-6282-4d82-b9f5-6b9587f7d805",
          "title": "You are an Associate at an investment banking firm. Using the provided historical financials for Salesforce (CRM-US), ru",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "06c284ef-6282-4d82-b9f5-6b9587f7d805",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/06c284ef-6282-4d82-b9f5-6b9587f7d805.json",
            "docs/eval/fresh-room/FR-020/tasks/06c284ef-6282-4d82-b9f5-6b9587f7d805/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "8182334a-2a3b-4607-91f3-59d5b94f5cb0",
          "title": "You are an Associate at an investment banking firm. Outline the historical financials for Salesforce (CRM-US) from the I",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "8182334a-2a3b-4607-91f3-59d5b94f5cb0",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/8182334a-2a3b-4607-91f3-59d5b94f5cb0.json",
            "docs/eval/fresh-room/FR-020/tasks/8182334a-2a3b-4607-91f3-59d5b94f5cb0/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "1306dbd8-dec4-469e-b806-d09113dd1010",
          "title": "Hello. Please put together a cash flow sweep analysis for Deere & Company (ticker is NYSE: DE). This cash flow sweep is ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "1306dbd8-dec4-469e-b806-d09113dd1010",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/1306dbd8-dec4-469e-b806-d09113dd1010.json",
            "docs/eval/fresh-room/FR-020/tasks/1306dbd8-dec4-469e-b806-d09113dd1010/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "eb58ec7c-f236-47a6-8ca7-d7297f9187f6",
          "title": "I’m an investment banking analyst in the TMT team at XYZ Bank, and my VP has asked me to pull together a quick margin an",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "eb58ec7c-f236-47a6-8ca7-d7297f9187f6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/eb58ec7c-f236-47a6-8ca7-d7297f9187f6.json",
            "docs/eval/fresh-room/FR-020/tasks/eb58ec7c-f236-47a6-8ca7-d7297f9187f6/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "07727295-e199-4039-a0ee-582af96749cf",
          "title": "It is November 2025 and you are an investment banking analyst in the Consumer Coverage group at XYZ Bank, focused on bea",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "07727295-e199-4039-a0ee-582af96749cf",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/07727295-e199-4039-a0ee-582af96749cf.json",
            "docs/eval/fresh-room/FR-020/tasks/07727295-e199-4039-a0ee-582af96749cf/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "69fcdefe-f4af-4676-ba21-b5c23c6b2f7c",
          "title": "You are an Associate at an investment banking firm. Your VP has notified you of an upcoming meeting with Okta for a valu",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "69fcdefe-f4af-4676-ba21-b5c23c6b2f7c",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/69fcdefe-f4af-4676-ba21-b5c23c6b2f7c.json",
            "docs/eval/fresh-room/FR-020/tasks/69fcdefe-f4af-4676-ba21-b5c23c6b2f7c/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "9ed5fdf8-7ffa-4c26-92d8-e1f28308af52",
          "title": "I’m an investment banking analyst at XYZ bank in the TMT team, and my MD has asked me to pull together a clean relative-",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "9ed5fdf8-7ffa-4c26-92d8-e1f28308af52",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/9ed5fdf8-7ffa-4c26-92d8-e1f28308af52.json",
            "docs/eval/fresh-room/FR-020/tasks/9ed5fdf8-7ffa-4c26-92d8-e1f28308af52/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "fe996540-6e06-4a84-82d9-06d0f32f6cd9",
          "title": "I’m an investment banking analyst in the TMT team at XYZ Bank, and my Director asked me to put together a clean product-",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "fe996540-6e06-4a84-82d9-06d0f32f6cd9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/fe996540-6e06-4a84-82d9-06d0f32f6cd9.json",
            "docs/eval/fresh-room/FR-020/tasks/fe996540-6e06-4a84-82d9-06d0f32f6cd9/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "17d8c86f-487e-4f22-89ba-1f7789ff940f",
          "title": "I have been asked to put together a valuation comp set for a software company and to show a peer set of high growth soft",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "17d8c86f-487e-4f22-89ba-1f7789ff940f",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/17d8c86f-487e-4f22-89ba-1f7789ff940f.json",
            "docs/eval/fresh-room/FR-020/tasks/17d8c86f-487e-4f22-89ba-1f7789ff940f/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "6a08ca87-feb5-4061-b532-26e89c888951",
          "title": "You are an investment banking Associate at a top -tier bulge bracket firm. Your VP has asked you to prepare an updated t",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "6a08ca87-feb5-4061-b532-26e89c888951",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/6a08ca87-feb5-4061-b532-26e89c888951.json",
            "docs/eval/fresh-room/FR-020/tasks/6a08ca87-feb5-4061-b532-26e89c888951/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "f0bd3730-d826-46f2-8422-b48480fa6250",
          "title": "Your MD was approached by a strategic / Corporate looking into potentially acquiring the New York Times (Ticker: NYT). T",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "f0bd3730-d826-46f2-8422-b48480fa6250",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/f0bd3730-d826-46f2-8422-b48480fa6250.json",
            "docs/eval/fresh-room/FR-020/tasks/f0bd3730-d826-46f2-8422-b48480fa6250/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "bd2aa5d6-30cd-4b6f-875a-2c3626f56441",
          "title": "Your MD was approached by a private equity client asking for the latest M&A precedents for companies in the tech and med",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "bd2aa5d6-30cd-4b6f-875a-2c3626f56441",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/bd2aa5d6-30cd-4b6f-875a-2c3626f56441.json",
            "docs/eval/fresh-room/FR-020/tasks/bd2aa5d6-30cd-4b6f-875a-2c3626f56441/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "60ed36b0-5e15-4c2d-b4a9-0f3d3cab9b16",
          "title": "I’m an investment banking analyst at XYZ bank working in the Technology group. Our MD has asked me to create a trading c",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "60ed36b0-5e15-4c2d-b4a9-0f3d3cab9b16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/60ed36b0-5e15-4c2d-b4a9-0f3d3cab9b16.json",
            "docs/eval/fresh-room/FR-020/tasks/60ed36b0-5e15-4c2d-b4a9-0f3d3cab9b16/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "432c6197-eade-4dcd-ae6d-0a30ec4da18c",
          "title": "You have been asked to show the stock price performance for select software companies (ADBE, AMZN, ORCL, GOOGL, NVDA, CR",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "432c6197-eade-4dcd-ae6d-0a30ec4da18c",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/432c6197-eade-4dcd-ae6d-0a30ec4da18c.json",
            "docs/eval/fresh-room/FR-020/tasks/432c6197-eade-4dcd-ae6d-0a30ec4da18c/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "df4c5a14-a6f9-4a56-ac8a-ad2562b50a7e",
          "title": "I'm an investment banking analyst at XYZ bank in the TMT team, and my VP has asked me to prepare a concise Q3 2025 earni",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "df4c5a14-a6f9-4a56-ac8a-ad2562b50a7e",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/df4c5a14-a6f9-4a56-ac8a-ad2562b50a7e.json",
            "docs/eval/fresh-room/FR-020/tasks/df4c5a14-a6f9-4a56-ac8a-ad2562b50a7e/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "5495a5a9-1087-4fd9-97f9-d677dd856e24",
          "title": "Your client is a diversified large-cap medtech company looking to acquire Insulet, a single product insulin pump manufac",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "5495a5a9-1087-4fd9-97f9-d677dd856e24",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/5495a5a9-1087-4fd9-97f9-d677dd856e24.json",
            "docs/eval/fresh-room/FR-020/tasks/5495a5a9-1087-4fd9-97f9-d677dd856e24/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "6b25d3dc-4b81-4d2e-b153-466b17040aae",
          "title": "Role: Act as an Investment Banking Analyst in the M&A / LevFin product group advising Aurora Private Equity on a buy-sid",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "6b25d3dc-4b81-4d2e-b153-466b17040aae",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/6b25d3dc-4b81-4d2e-b153-466b17040aae.json",
            "docs/eval/fresh-room/FR-020/tasks/6b25d3dc-4b81-4d2e-b153-466b17040aae/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "98bdf8f3-98b7-4d62-ac39-5cee23623d83",
          "title": "Your MD is asking you to put together a two page profile on Meta, to pitch a potential buyside client on acquiring a min",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "98bdf8f3-98b7-4d62-ac39-5cee23623d83",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/98bdf8f3-98b7-4d62-ac39-5cee23623d83.json",
            "docs/eval/fresh-room/FR-020/tasks/98bdf8f3-98b7-4d62-ac39-5cee23623d83/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "205a3cb3-33b8-42ea-b50c-fe2dc1630208",
          "title": "I’m an investment banking analyst at XYZ bank in the TMT team, and my VP has asked me to pull together a quick overview ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "205a3cb3-33b8-42ea-b50c-fe2dc1630208",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/205a3cb3-33b8-42ea-b50c-fe2dc1630208.json",
            "docs/eval/fresh-room/FR-020/tasks/205a3cb3-33b8-42ea-b50c-fe2dc1630208/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "f1c210d9-6357-4fad-828e-252e6f6d59d3",
          "title": "I’m an investment banking analyst in the TMT team at XYZ Bank, and my MD has asked me to pull together a long-term subsc",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "f1c210d9-6357-4fad-828e-252e6f6d59d3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/f1c210d9-6357-4fad-828e-252e6f6d59d3.json",
            "docs/eval/fresh-room/FR-020/tasks/f1c210d9-6357-4fad-828e-252e6f6d59d3/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "19b3361c-82c3-41d6-b941-d3f60ce32026",
          "title": "You are on an M&A project and have just successfully won the mandate for a sell side following a pitch. Your job is to c",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "19b3361c-82c3-41d6-b941-d3f60ce32026",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/19b3361c-82c3-41d6-b941-d3f60ce32026.json",
            "docs/eval/fresh-room/FR-020/tasks/19b3361c-82c3-41d6-b941-d3f60ce32026/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "963978db-b155-4fe2-9ae7-ca0b259a5fba",
          "title": "Your MD has asked you to prepare a deep dive company overview/company profile on Solarwinds (a public company). Prepare ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "963978db-b155-4fe2-9ae7-ca0b259a5fba",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/963978db-b155-4fe2-9ae7-ca0b259a5fba.json",
            "docs/eval/fresh-room/FR-020/tasks/963978db-b155-4fe2-9ae7-ca0b259a5fba/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "4fb6232d-c2a9-4141-98de-3e2025bdf52b",
          "title": "Role: You are an Investment Banking Analyst in the M&A Group advising Shipwell (the Client) on a potential sale.\n\nContex",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "4fb6232d-c2a9-4141-98de-3e2025bdf52b",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/4fb6232d-c2a9-4141-98de-3e2025bdf52b.json",
            "docs/eval/fresh-room/FR-020/tasks/4fb6232d-c2a9-4141-98de-3e2025bdf52b/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "096a6840-718c-4ab4-8dff-10601ffe003c",
          "title": "Our client is Sonoco Products Company (NYSE: SON), a global packaging firm that is looking to sell its asset ThermoSafe ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "096a6840-718c-4ab4-8dff-10601ffe003c",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/096a6840-718c-4ab4-8dff-10601ffe003c.json",
            "docs/eval/fresh-room/FR-020/tasks/096a6840-718c-4ab4-8dff-10601ffe003c/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c1c36cfd-677b-462b-96d4-ea76e12e98e8",
          "title": "Your VP has asked you to prepare a refreshed company profile for Adobe (NASDAQ: ADBE) for use in external discussions wi",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c1c36cfd-677b-462b-96d4-ea76e12e98e8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c1c36cfd-677b-462b-96d4-ea76e12e98e8.json",
            "docs/eval/fresh-room/FR-020/tasks/c1c36cfd-677b-462b-96d4-ea76e12e98e8/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "69507fd6-a925-4e1a-8fc3-6485c71dc828",
          "title": "Draft the products and services section of a Confidential Information Memorandum for the Greenbrier Company, Inc.",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "69507fd6-a925-4e1a-8fc3-6485c71dc828",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/69507fd6-a925-4e1a-8fc3-6485c71dc828.json",
            "docs/eval/fresh-room/FR-020/tasks/69507fd6-a925-4e1a-8fc3-6485c71dc828/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "1b181d77-ad4c-498a-b76e-c8503c05b7b7",
          "title": "Please deliver a single-slide PPT and its corresponding PDF that includes a clean terms summary of Bank of America Corpo",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "1b181d77-ad4c-498a-b76e-c8503c05b7b7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/1b181d77-ad4c-498a-b76e-c8503c05b7b7.json",
            "docs/eval/fresh-room/FR-020/tasks/1b181d77-ad4c-498a-b76e-c8503c05b7b7/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "515a104d-ab72-47fe-b1a2-d67325ba6473",
          "title": "\"It is 3/3/2025, and you are a leveraged finance associate at a bulge-bracket investment bank. A group of private equity",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "515a104d-ab72-47fe-b1a2-d67325ba6473",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/515a104d-ab72-47fe-b1a2-d67325ba6473.json",
            "docs/eval/fresh-room/FR-020/tasks/515a104d-ab72-47fe-b1a2-d67325ba6473/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "eb2f7b9c-2f96-4093-bdd5-e28707e707a1",
          "title": "You are an analyst at a global private equity (PE) firm building a leveraged buyout (LBO) valuation model.\nThe valuation",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "eb2f7b9c-2f96-4093-bdd5-e28707e707a1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/eb2f7b9c-2f96-4093-bdd5-e28707e707a1.json",
            "docs/eval/fresh-room/FR-020/tasks/eb2f7b9c-2f96-4093-bdd5-e28707e707a1/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "8eb06eaf-d2ce-4f86-9239-d4a10bc72cf6",
          "title": "A private equity firm is evaluating taking CAKE (The Cheesecake Factory) private. The MD is asking you to build a full L",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "8eb06eaf-d2ce-4f86-9239-d4a10bc72cf6",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/8eb06eaf-d2ce-4f86-9239-d4a10bc72cf6.json",
            "docs/eval/fresh-room/FR-020/tasks/8eb06eaf-d2ce-4f86-9239-d4a10bc72cf6/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "bc55d3e8-0d59-43d6-a385-03c5330e47bf",
          "title": "You are an investment banking analyst in the Technology M&A group tasked with evaluating the intrinsic value of Uber Tec",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "bc55d3e8-0d59-43d6-a385-03c5330e47bf",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/bc55d3e8-0d59-43d6-a385-03c5330e47bf.json",
            "docs/eval/fresh-room/FR-020/tasks/bc55d3e8-0d59-43d6-a385-03c5330e47bf/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "1d073c85-42fe-4d6d-8539-069ea87082f8",
          "title": "You are a leveraged finance analyst at a bulge-bracket investment bank. A private equity firm is planning to take Shake ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "1d073c85-42fe-4d6d-8539-069ea87082f8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/1d073c85-42fe-4d6d-8539-069ea87082f8.json",
            "docs/eval/fresh-room/FR-020/tasks/1d073c85-42fe-4d6d-8539-069ea87082f8/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "a0b2858a-1ffd-423d-8948-e14aceaa1f69",
          "title": "You are an M&A analyst, and your MD has tasked you with advising a strategic buyer on a potential acquisition by buildin",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "a0b2858a-1ffd-423d-8948-e14aceaa1f69",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/a0b2858a-1ffd-423d-8948-e14aceaa1f69.json",
            "docs/eval/fresh-room/FR-020/tasks/a0b2858a-1ffd-423d-8948-e14aceaa1f69/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "5d72295c-f640-4c78-b85a-6f338e46c02b",
          "title": "You are a leverage finance analyst at a bulge-bracket bank. A consortium of private equity firms is taking Boeing privat",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "5d72295c-f640-4c78-b85a-6f338e46c02b",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/5d72295c-f640-4c78-b85a-6f338e46c02b.json",
            "docs/eval/fresh-room/FR-020/tasks/5d72295c-f640-4c78-b85a-6f338e46c02b/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "7b078eaa-4fe4-4780-9f53-531872b73274",
          "title": "You are an analyst in the M&A group at a bulge-bracket bank. Given the geopolitical landscape and the expected tailwinds",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "7b078eaa-4fe4-4780-9f53-531872b73274",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/7b078eaa-4fe4-4780-9f53-531872b73274.json",
            "docs/eval/fresh-room/FR-020/tasks/7b078eaa-4fe4-4780-9f53-531872b73274/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "e3c47ea1-cc6c-4981-9943-bc85162fc2a8",
          "title": "You are first year analyst in the consumer and retail team and are asked by your MD to do a quick preliminary DCF analys",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "e3c47ea1-cc6c-4981-9943-bc85162fc2a8",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/e3c47ea1-cc6c-4981-9943-bc85162fc2a8.json",
            "docs/eval/fresh-room/FR-020/tasks/e3c47ea1-cc6c-4981-9943-bc85162fc2a8/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "44bd7a6b-b3c3-4eb8-b56d-6a769bdbf58e",
          "title": "The CFO of Papa John’s Pizza believes that the current footprint modernization project is being funded in a suboptimal m",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "44bd7a6b-b3c3-4eb8-b56d-6a769bdbf58e",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/44bd7a6b-b3c3-4eb8-b56d-6a769bdbf58e.json",
            "docs/eval/fresh-room/FR-020/tasks/44bd7a6b-b3c3-4eb8-b56d-6a769bdbf58e/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "b1b4a2a3-6d5b-441f-a17e-41753656a0e3",
          "title": "You are an Investment Banking Analyst tasked with building a three-statement operating model for Affirm Holdings, Inc. (",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "b1b4a2a3-6d5b-441f-a17e-41753656a0e3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/b1b4a2a3-6d5b-441f-a17e-41753656a0e3.json",
            "docs/eval/fresh-room/FR-020/tasks/b1b4a2a3-6d5b-441f-a17e-41753656a0e3/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "6801c651-fd1b-4c12-a4ce-1338442c7ab1",
          "title": "It is 01/28/2026. You are an investment banking analyst at a global bank. Your managing director wants to assess what th",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "6801c651-fd1b-4c12-a4ce-1338442c7ab1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/6801c651-fd1b-4c12-a4ce-1338442c7ab1.json",
            "docs/eval/fresh-room/FR-020/tasks/6801c651-fd1b-4c12-a4ce-1338442c7ab1/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "aa3e4d0f-f456-423b-893b-9d4e88c64e74",
          "title": "I am an investment banking analyst in the REGAL coverage group at XYZ bank. I'm covering Real Estate, Gaming and Lodging",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "aa3e4d0f-f456-423b-893b-9d4e88c64e74",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/aa3e4d0f-f456-423b-893b-9d4e88c64e74.json",
            "docs/eval/fresh-room/FR-020/tasks/aa3e4d0f-f456-423b-893b-9d4e88c64e74/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "433c9e48-4b74-49a4-8d2b-d8c25f445c60",
          "title": "You are an analyst in the M&A group at XYZ Bank. Your MD has been in discussion with Sponsor ABC about taking the Cheese",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "433c9e48-4b74-49a4-8d2b-d8c25f445c60",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/433c9e48-4b74-49a4-8d2b-d8c25f445c60.json",
            "docs/eval/fresh-room/FR-020/tasks/433c9e48-4b74-49a4-8d2b-d8c25f445c60/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "eeba6325-7212-4924-8760-91065d3b1af3",
          "title": "I am an investment banking analyst in the TMT group at XYZ bank. My team has been engaged by a financial sponsor evaluat",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "eeba6325-7212-4924-8760-91065d3b1af3",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/eeba6325-7212-4924-8760-91065d3b1af3.json",
            "docs/eval/fresh-room/FR-020/tasks/eeba6325-7212-4924-8760-91065d3b1af3/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "7638f425-6102-4ab9-ae75-ba88c0b8fb8d",
          "title": "You are a leverage finance analyst at ABC bank. An MD from your group is in talks with a private equity company to take ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "7638f425-6102-4ab9-ae75-ba88c0b8fb8d",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/7638f425-6102-4ab9-ae75-ba88c0b8fb8d.json",
            "docs/eval/fresh-room/FR-020/tasks/7638f425-6102-4ab9-ae75-ba88c0b8fb8d/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "93c1e85c-7954-4464-8ff3-4c2c920cad89",
          "title": "You are an analyst in the LevFin Industrial pod working on potential financing opportunities for Entegris. Your MD has a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "93c1e85c-7954-4464-8ff3-4c2c920cad89",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/93c1e85c-7954-4464-8ff3-4c2c920cad89.json",
            "docs/eval/fresh-room/FR-020/tasks/93c1e85c-7954-4464-8ff3-4c2c920cad89/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "067cb834-ad43-448b-a523-d0c9898656e0",
          "title": "You are an investment banking associate working on the M&A team focused on the Communications sector, and your MD has as",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "067cb834-ad43-448b-a523-d0c9898656e0",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/067cb834-ad43-448b-a523-d0c9898656e0.json",
            "docs/eval/fresh-room/FR-020/tasks/067cb834-ad43-448b-a523-d0c9898656e0/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "68259d37-5dd5-4c19-bafc-676246a479ea",
          "title": "Please produce a slide showing post-IPO trading performance for Cava Group, Inc. (ticker: CAVA) for your Equity Capital ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "68259d37-5dd5-4c19-bafc-676246a479ea",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/68259d37-5dd5-4c19-bafc-676246a479ea.json",
            "docs/eval/fresh-room/FR-020/tasks/68259d37-5dd5-4c19-bafc-676246a479ea/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "d2c04408-4c3b-45a6-a076-7f5ccca8fc2a",
          "title": "You are a 2nd-year banker in the M&A group at a bulge-bracket investment bank. Your MD just let you know that your team ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "d2c04408-4c3b-45a6-a076-7f5ccca8fc2a",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/d2c04408-4c3b-45a6-a076-7f5ccca8fc2a.json",
            "docs/eval/fresh-room/FR-020/tasks/d2c04408-4c3b-45a6-a076-7f5ccca8fc2a/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "f8571fea-f310-4a46-be0c-bbfb65990198",
          "title": "I am an investment banking analyst in the Consumer Retail group at XYZ Bank. My VP has several meetings coming up with C",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "f8571fea-f310-4a46-be0c-bbfb65990198",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/f8571fea-f310-4a46-be0c-bbfb65990198.json",
            "docs/eval/fresh-room/FR-020/tasks/f8571fea-f310-4a46-be0c-bbfb65990198/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "3c70cabb-2915-4d5f-8e62-cd6b729a511f",
          "title": "You're a first-year Analyst in the Tech coverage investment banking group. You have been tasked with helping a client, D",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "3c70cabb-2915-4d5f-8e62-cd6b729a511f",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/3c70cabb-2915-4d5f-8e62-cd6b729a511f.json",
            "docs/eval/fresh-room/FR-020/tasks/3c70cabb-2915-4d5f-8e62-cd6b729a511f/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "40097508-c68a-4066-9031-e99eac4f7a18",
          "title": "Build a fully integrated leverage finance model which is a buyout Sources & Uses model with debt structuring for JetBlue",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "40097508-c68a-4066-9031-e99eac4f7a18",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/40097508-c68a-4066-9031-e99eac4f7a18.json",
            "docs/eval/fresh-room/FR-020/tasks/40097508-c68a-4066-9031-e99eac4f7a18/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "42449f25-042c-4260-8a68-3c9c35009501",
          "title": "You work in Private Equity and you must build a Leveraged Buyout model for a veterinary pharmacy. The company is to be a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "42449f25-042c-4260-8a68-3c9c35009501",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/42449f25-042c-4260-8a68-3c9c35009501.json",
            "docs/eval/fresh-room/FR-020/tasks/42449f25-042c-4260-8a68-3c9c35009501/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "507a3d72-91bd-4e0c-9c77-c0cf521f4a71",
          "title": "You are a Private Equity Associate at PE Partners, a middle-market consumer products fund. Your MD wants you to build an",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "507a3d72-91bd-4e0c-9c77-c0cf521f4a71",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/507a3d72-91bd-4e0c-9c77-c0cf521f4a71.json",
            "docs/eval/fresh-room/FR-020/tasks/507a3d72-91bd-4e0c-9c77-c0cf521f4a71/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "8bdef554-4723-42ff-a813-4711f9b330af",
          "title": "Today is December 31, 2025. You are an associate at a middle market buyout fund evaluating a potential acquisition of a ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "8bdef554-4723-42ff-a813-4711f9b330af",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/8bdef554-4723-42ff-a813-4711f9b330af.json",
            "docs/eval/fresh-room/FR-020/tasks/8bdef554-4723-42ff-a813-4711f9b330af/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "907e3e2e-c733-47bf-bedc-3650147bbe91",
          "title": "You are an Investment Banking Analyst tasked with building a discounted cash flow valuation model for YUM! Brands, Inc. ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "907e3e2e-c733-47bf-bedc-3650147bbe91",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/907e3e2e-c733-47bf-bedc-3650147bbe91.json",
            "docs/eval/fresh-room/FR-020/tasks/907e3e2e-c733-47bf-bedc-3650147bbe91/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "92235649-0d63-4396-b864-a64257e16746",
          "title": "Help me put together a view on trading multiples for online travel so we can anchor a valuation range – I want to unders",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "92235649-0d63-4396-b864-a64257e16746",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/92235649-0d63-4396-b864-a64257e16746.json",
            "docs/eval/fresh-room/FR-020/tasks/92235649-0d63-4396-b864-a64257e16746/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "aa577c1d-2c22-4963-91c4-456ac32e5b4a",
          "title": "It is January 28, 2026, and you are an analyst on the M&A team at a bulge bracket bank. Your VP has asked you to build a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "aa577c1d-2c22-4963-91c4-456ac32e5b4a",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/aa577c1d-2c22-4963-91c4-456ac32e5b4a.json",
            "docs/eval/fresh-room/FR-020/tasks/aa577c1d-2c22-4963-91c4-456ac32e5b4a/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "ac1a3956-b087-446b-9717-0f3d529e8fff",
          "title": "You are an investment banker at xyz bank who is covering a sponsor in the coverage space you are in. Your MD has a spons",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "ac1a3956-b087-446b-9717-0f3d529e8fff",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/ac1a3956-b087-446b-9717-0f3d529e8fff.json",
            "docs/eval/fresh-room/FR-020/tasks/ac1a3956-b087-446b-9717-0f3d529e8fff/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c6926dfc-b844-40f6-b530-2198b39d8ddd",
          "title": "I’d like you to build a clean, fully dynamic equity waterfall model in Excel that calculates sponsor, rollover equity (s",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c6926dfc-b844-40f6-b530-2198b39d8ddd",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c6926dfc-b844-40f6-b530-2198b39d8ddd.json",
            "docs/eval/fresh-room/FR-020/tasks/c6926dfc-b844-40f6-b530-2198b39d8ddd/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c7bbdb93-8b29-44f6-8e5f-13fc0d3de8fc",
          "title": "Can you please build a clean, investor-ready deliverable titled “Project Redwood – Pro Forma Capitalization & Ownership ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c7bbdb93-8b29-44f6-8e5f-13fc0d3de8fc",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c7bbdb93-8b29-44f6-8e5f-13fc0d3de8fc.json",
            "docs/eval/fresh-room/FR-020/tasks/c7bbdb93-8b29-44f6-8e5f-13fc0d3de8fc/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "2048736b-7bb8-4c54-899e-ddfdf12641ab",
          "title": "You’re an Investment Banking Analyst working in the DCM space, and your MD has tasked you with preparing a Rating Agency",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "2048736b-7bb8-4c54-899e-ddfdf12641ab",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/2048736b-7bb8-4c54-899e-ddfdf12641ab.json",
            "docs/eval/fresh-room/FR-020/tasks/2048736b-7bb8-4c54-899e-ddfdf12641ab/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "4194bf8e-eede-480c-b012-b68ab71837c1",
          "title": "Please build a clean, investor-ready slide titled “Project Tiburon – Cap Table Dilution Analysis” that shows the full re",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "4194bf8e-eede-480c-b012-b68ab71837c1",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/4194bf8e-eede-480c-b012-b68ab71837c1.json",
            "docs/eval/fresh-room/FR-020/tasks/4194bf8e-eede-480c-b012-b68ab71837c1/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "ed5ac3f0-a9c5-4daa-b97b-e5acc18ae423",
          "title": "We're going into a private equity client's office on Wednesday to pitch for the sale of one of their portfolio companies",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "ed5ac3f0-a9c5-4daa-b97b-e5acc18ae423",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/ed5ac3f0-a9c5-4daa-b97b-e5acc18ae423.json",
            "docs/eval/fresh-room/FR-020/tasks/ed5ac3f0-a9c5-4daa-b97b-e5acc18ae423/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c032d3c9-5055-4346-a60d-429f658f71f9",
          "title": "Ahead of the group’s upcoming meeting with the company, your VP has asked you to run a trading comps analysis for Pepsi ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c032d3c9-5055-4346-a60d-429f658f71f9",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c032d3c9-5055-4346-a60d-429f658f71f9.json",
            "docs/eval/fresh-room/FR-020/tasks/c032d3c9-5055-4346-a60d-429f658f71f9/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "0fc7bc3c-f10b-48f1-acee-1db24a4f75fc",
          "title": "You are building a one-page PowerPoint teaser that is 8 1/2 x 11\" and we need to present the information of Zoom (Ticker",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "0fc7bc3c-f10b-48f1-acee-1db24a4f75fc",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/0fc7bc3c-f10b-48f1-acee-1db24a4f75fc.json",
            "docs/eval/fresh-room/FR-020/tasks/0fc7bc3c-f10b-48f1-acee-1db24a4f75fc/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "daa6b204-9dc9-491f-bdf9-963208578db2",
          "title": "I'm an Investment Banking Analyst at XYZ Bank in their Transportation & Logistics Group. My VP has asked me to pull toge",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "daa6b204-9dc9-491f-bdf9-963208578db2",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/daa6b204-9dc9-491f-bdf9-963208578db2.json",
            "docs/eval/fresh-room/FR-020/tasks/daa6b204-9dc9-491f-bdf9-963208578db2/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "5f1e6f38-3380-43f4-a233-b758706985b7",
          "title": "You are a first-year investment banking analyst in the Energy group at a bulge bracket firm preparing a trading comparab",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "5f1e6f38-3380-43f4-a233-b758706985b7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/5f1e6f38-3380-43f4-a233-b758706985b7.json",
            "docs/eval/fresh-room/FR-020/tasks/5f1e6f38-3380-43f4-a233-b758706985b7/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "960dc224-634f-435b-a42d-f8edac8ddd2f",
          "title": "Hello. Please build a DCF model for Keurig Dr Pepper (KDP) using 2021-2024 historicals and a 5-year projection period (2",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "960dc224-634f-435b-a42d-f8edac8ddd2f",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/960dc224-634f-435b-a42d-f8edac8ddd2f.json",
            "docs/eval/fresh-room/FR-020/tasks/960dc224-634f-435b-a42d-f8edac8ddd2f/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "f205ac8c-4dae-4620-b2c9-ccc44858ec70",
          "title": "You are an associate at a private equity firm. Your managing director has asked you to build a leveraged buyout model fo",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "f205ac8c-4dae-4620-b2c9-ccc44858ec70",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/f205ac8c-4dae-4620-b2c9-ccc44858ec70.json",
            "docs/eval/fresh-room/FR-020/tasks/f205ac8c-4dae-4620-b2c9-ccc44858ec70/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "c292330e-d88f-4353-88fd-6b0f8b3b93c2",
          "title": "Build a DCF model for Chipotle Mexican Grill (CMG) with 4 years of historicals (2021–2024) and 5 years of projections (2",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "c292330e-d88f-4353-88fd-6b0f8b3b93c2",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/c292330e-d88f-4353-88fd-6b0f8b3b93c2.json",
            "docs/eval/fresh-room/FR-020/tasks/c292330e-d88f-4353-88fd-6b0f8b3b93c2/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "8a19d926-4db7-4068-aabd-6b34bf965f16",
          "title": "Build an integrated three-statement operating model (Income Statement, Balance Sheet, Cash Flow Statement) with a debt s",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "8a19d926-4db7-4068-aabd-6b34bf965f16",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/8a19d926-4db7-4068-aabd-6b34bf965f16.json",
            "docs/eval/fresh-room/FR-020/tasks/8a19d926-4db7-4068-aabd-6b34bf965f16/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "50f1b669-f6c3-453f-9d80-8aee2b1e01af",
          "title": "You are a DCM investment banking analyst preparing an internal credit snapshot for Verizon Communications Inc. (Ticker: ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "50f1b669-f6c3-453f-9d80-8aee2b1e01af",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/50f1b669-f6c3-453f-9d80-8aee2b1e01af.json",
            "docs/eval/fresh-room/FR-020/tasks/50f1b669-f6c3-453f-9d80-8aee2b1e01af/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "8e61629c-5c68-4ff4-b680-222b56b04f55",
          "title": "Ahead of our coverage call with the UPS treasury team, I need a secondary market monitoring sheet on their outstanding U",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "8e61629c-5c68-4ff4-b680-222b56b04f55",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/8e61629c-5c68-4ff4-b680-222b56b04f55.json",
            "docs/eval/fresh-room/FR-020/tasks/8e61629c-5c68-4ff4-b680-222b56b04f55/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "cf5884e8-8881-4747-a65a-10d8b468a9f4",
          "title": "You are an analyst at XYZ bank on the consumer M&A team. The team’s top client has been tracking ETSY (ETSY-US) as a pot",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "cf5884e8-8881-4747-a65a-10d8b468a9f4",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/cf5884e8-8881-4747-a65a-10d8b468a9f4.json",
            "docs/eval/fresh-room/FR-020/tasks/cf5884e8-8881-4747-a65a-10d8b468a9f4/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "b486dc21-d67e-4857-b2f9-b78c0e0222ce",
          "title": "Take a shot at framing the valuation landscape for the United Therapeutics board; we need to show them exactly where the",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "b486dc21-d67e-4857-b2f9-b78c0e0222ce",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/b486dc21-d67e-4857-b2f9-b78c0e0222ce.json",
            "docs/eval/fresh-room/FR-020/tasks/b486dc21-d67e-4857-b2f9-b78c0e0222ce/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "55fd995a-e4df-4942-aee1-ee1be9a72b6a",
          "title": "You are a biotechnology investment banking analyst covering mid cap clinical stage companies. Your Managing Director has",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "55fd995a-e4df-4942-aee1-ee1be9a72b6a",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/55fd995a-e4df-4942-aee1-ee1be9a72b6a.json",
            "docs/eval/fresh-room/FR-020/tasks/55fd995a-e4df-4942-aee1-ee1be9a72b6a/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "cc4c8473-4c54-4fdd-9af8-c55bc9e4b07d",
          "title": "It is July 31st, 2025. Figma, Inc. has just priced its IPO and filed its 424B4 pricing prospectus. You are an investment",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "cc4c8473-4c54-4fdd-9af8-c55bc9e4b07d",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/cc4c8473-4c54-4fdd-9af8-c55bc9e4b07d.json",
            "docs/eval/fresh-room/FR-020/tasks/cc4c8473-4c54-4fdd-9af8-c55bc9e4b07d/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "73c10eda-38d8-4b48-bac6-baaaff3d937f",
          "title": "You are an Investment Banking Analyst tasked with building a fully structured Analysis at Various Prices (AVP) analysis ",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "73c10eda-38d8-4b48-bac6-baaaff3d937f",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/73c10eda-38d8-4b48-bac6-baaaff3d937f.json",
            "docs/eval/fresh-room/FR-020/tasks/73c10eda-38d8-4b48-bac6-baaaff3d937f/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "31f70ac1-01c3-4971-84f8-0a2438fb5b92",
          "title": "DELIVERABLE:\nCreate a 2-slide PowerPoint presentation (exported to PDF) on the Handshake Advisory template covering:\n\nSl",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "31f70ac1-01c3-4971-84f8-0a2438fb5b92",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/31f70ac1-01c3-4971-84f8-0a2438fb5b92.json",
            "docs/eval/fresh-room/FR-020/tasks/31f70ac1-01c3-4971-84f8-0a2438fb5b92/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "1b253d04-2d32-41e8-a0b5-5450a0808e3b",
          "title": "Prepare a concise, institutionally rigorous buyer universe presentation assessing potential acquirers of Transco from Th",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "1b253d04-2d32-41e8-a0b5-5450a0808e3b",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/1b253d04-2d32-41e8-a0b5-5450a0808e3b.json",
            "docs/eval/fresh-room/FR-020/tasks/1b253d04-2d32-41e8-a0b5-5450a0808e3b/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "7d11ce66-d850-4606-a3ef-f3960db34df7",
          "title": "You are an Analyst in the TMT M&A group at bank XYZ. You are tasked by your VP to build a merger model for a potential a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "7d11ce66-d850-4606-a3ef-f3960db34df7",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/7d11ce66-d850-4606-a3ef-f3960db34df7.json",
            "docs/eval/fresh-room/FR-020/tasks/7d11ce66-d850-4606-a3ef-f3960db34df7/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "41e130fe-89f0-4b2b-97ee-d727e1fa0afd",
          "title": "You are a second-year investment banking analyst supporting a valuation workstream for Amazon (AMZN) in the context of a",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "41e130fe-89f0-4b2b-97ee-d727e1fa0afd",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/41e130fe-89f0-4b2b-97ee-d727e1fa0afd.json",
            "docs/eval/fresh-room/FR-020/tasks/41e130fe-89f0-4b2b-97ee-d727e1fa0afd/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "4fedc1aa-5e19-4339-9f0b-f455239b2655",
          "title": "You are an investment banking analyst in the Consumer & Retail group of an investment bank. Your associate has asked you",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "4fedc1aa-5e19-4339-9f0b-f455239b2655",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/4fedc1aa-5e19-4339-9f0b-f455239b2655.json",
            "docs/eval/fresh-room/FR-020/tasks/4fedc1aa-5e19-4339-9f0b-f455239b2655/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "6a2f5cf8-be91-4801-87b8-226f259369fb",
          "title": "It is 11/28/2025. You are an investment banking analyst at [Bank]. An institutional investor in Amazon (AMZN) is seeking",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "6a2f5cf8-be91-4801-87b8-226f259369fb",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/6a2f5cf8-be91-4801-87b8-226f259369fb.json",
            "docs/eval/fresh-room/FR-020/tasks/6a2f5cf8-be91-4801-87b8-226f259369fb/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "73450a4d-5954-496c-a9f8-17f712bc5521",
          "title": "Build a fully integrated DCF valuation model for Essential Utilities (Ticker: WTRG) as of 12/31/2024. Calculate fully di",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "73450a4d-5954-496c-a9f8-17f712bc5521",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/73450a4d-5954-496c-a9f8-17f712bc5521.json",
            "docs/eval/fresh-room/FR-020/tasks/73450a4d-5954-496c-a9f8-17f712bc5521/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "ceba18d0-5558-4637-be3a-a97121796201",
          "title": "Your MD has asked you to run a DCF analysis for Coca-Cola Company (NYSE: KO). The output should be built in Excel and ti",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "ceba18d0-5558-4637-be3a-a97121796201",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/ceba18d0-5558-4637-be3a-a97121796201.json",
            "docs/eval/fresh-room/FR-020/tasks/ceba18d0-5558-4637-be3a-a97121796201/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        },
        {
          "familyId": "bankertoolbench-full-100",
          "taskId": "44e44478-9e36-4df9-9657-85797c7fee58",
          "title": "You are an investment banking analyst on the ECM desk. Your bank is lead-left bookrunner on a $350M IPO for NovaCrest Te",
          "status": "local_live_browser_only",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": true,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:btb",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "BTB_LIVE_ROOM_E2E": "1",
              "BTB_UI_BUNDLE_ROOT": ".tmp/official-benchmarks/bankertoolbench-repo/btb-data",
              "BTB_UI_TASK_ID": "44e44478-9e36-4df9-9657-85797c7fee58",
              "BENCH_AGENT_MODEL_MODE": "specific",
              "BENCH_AGENT_MODEL_POLICY": "z-ai/glm-5.2"
            }
          },
          "evidence": [
            ".tmp/official-benchmarks/bankertoolbench-repo/btb-data/tasks.jsonl",
            "docs/eval/bankertoolbench/live-room/44e44478-9e36-4df9-9657-85797c7fee58.json",
            "docs/eval/fresh-room/FR-020/tasks/44e44478-9e36-4df9-9657-85797c7fee58/latest.json"
          ],
          "blockers": [
            "Existing BTB receipt is local live-browser only; rerun against https://noderoom.live."
          ]
        }
      ]
    },
    {
      "id": "accounting-live-proofloop",
      "title": "Accounting live proof-loop",
      "taskCount": 4,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 4,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "accounting-live-proofloop",
          "taskId": "variance-calc",
          "title": "Q3 Variance Calculation",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:accounting:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/accounting/live.accounting.config.json",
              "PROOFLOOP_TASK_IDS": "variance-calc",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/accounting/live.accounting.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts",
            "docs/eval/proofloop-live-accounting-free-smoke.json",
            "https://noderoom.live/?room=PLMR7HIVFE&name=Proof+Loop",
            "receipt_family=accounting-live-proofloop",
            "receipt_task=variance-calc",
            "requested_model=qwen/qwen3-coder:free",
            "actual_model=qwen/qwen3-coder:free",
            "measured_cost_usd=0"
          ],
          "blockers": [
            "Committed receipt docs/eval/proofloop-live-accounting-free-smoke.json must set officialScoreClaim:false before it can count as proxy proof.",
            "Committed receipt docs/eval/proofloop-live-accounting-free-smoke.json did not pass overall.",
            "Committed receipt docs/eval/proofloop-live-accounting-free-smoke.json did not pass task variance-calc."
          ]
        },
        {
          "familyId": "accounting-live-proofloop",
          "taskId": "research-enrich",
          "title": "Company Research Enrichment",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:accounting:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/accounting/live.accounting.config.json",
              "PROOFLOOP_TASK_IDS": "research-enrich",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/accounting/live.accounting.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Accounting live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        },
        {
          "familyId": "accounting-live-proofloop",
          "taskId": "runway-calc",
          "title": "Runway Calculation",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:accounting:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/accounting/live.accounting.config.json",
              "PROOFLOOP_TASK_IDS": "runway-calc",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/accounting/live.accounting.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Accounting live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        },
        {
          "familyId": "accounting-live-proofloop",
          "taskId": "memo-draft",
          "title": "Diligence Memo Draft",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:accounting:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/accounting/live.accounting.config.json",
              "PROOFLOOP_TASK_IDS": "memo-draft",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/accounting/live.accounting.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Accounting live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        }
      ]
    },
    {
      "id": "notion-live-proofloop",
      "title": "Notion SDR/BDR live proof-loop",
      "taskCount": 4,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 4,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "notion-live-proofloop",
          "taskId": "company-research",
          "title": "Company Research — Mercury",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:notion:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/notion/live.notion.config.json",
              "PROOFLOOP_TASK_IDS": "company-research",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/notion/live.notion.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Notion live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        },
        {
          "familyId": "notion-live-proofloop",
          "taskId": "competitor-analysis",
          "title": "Competitor Analysis — Ramp vs Brex",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:notion:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/notion/live.notion.config.json",
              "PROOFLOOP_TASK_IDS": "competitor-analysis",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/notion/live.notion.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Notion live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        },
        {
          "familyId": "notion-live-proofloop",
          "taskId": "diligence-memo",
          "title": "Diligence Memo — CardioNova",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:notion:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/notion/live.notion.config.json",
              "PROOFLOOP_TASK_IDS": "diligence-memo",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/notion/live.notion.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Notion live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        },
        {
          "familyId": "notion-live-proofloop",
          "taskId": "workplan-update",
          "title": "Workplan Update",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:notion:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASKS_JSON": "proofloop/notion/live.notion.config.json",
              "PROOFLOOP_TASK_IDS": "workplan-update",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/notion/live.notion.config.json",
            "proofloop/live-browser-proof.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Notion live proof-loop has no passing prod live-browser receipt for this task/model yet."
          ]
        }
      ]
    },
    {
      "id": "proximitty-underwriting-pr0",
      "title": "Proximitty underwriting PR0",
      "taskCount": 4,
      "prodLiveBrowserPassed": 0,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 4,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "proximitty-underwriting-pr0",
          "taskId": "proximitty-intake",
          "title": "proximitty-intake.spec.ts",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:proximitty:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "proximitty-intake",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/scenarios/proximitty-intake.spec.ts",
            "proofloop/benchmarks/proximitty/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts",
            "docs/eval/proofloop-live-proximitty-free-smoke.json",
            "https://noderoom.live/?room=NRLQ2M1OG0P&name=Proof+Loop",
            "receipt_family=proximitty-underwriting-pr0",
            "receipt_task=proximitty-intake",
            "requested_model=qwen/qwen3-coder:free",
            "actual_model=z-ai/glm-4.7-flash",
            "measured_cost_usd=0.012"
          ],
          "blockers": [
            "model_route_mismatch: route_integrity=model_route_mismatch; requested=qwen/qwen3-coder:free; actual=z-ai/glm-4.7-flash; cost=$0.0120; failures=model_route_mismatch,free_route_used_paid_model,free_route_billed_nonzero_cost."
          ]
        },
        {
          "familyId": "proximitty-underwriting-pr0",
          "taskId": "proximitty-model-comparison",
          "title": "proximitty-model-comparison.spec.ts",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:proximitty:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "proximitty-model-comparison",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/scenarios/proximitty-model-comparison.spec.ts",
            "proofloop/benchmarks/proximitty/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Proximitty has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "proximitty-underwriting-pr0",
          "taskId": "proximitty-risk-research",
          "title": "proximitty-risk-research.spec.ts",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:proximitty:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "proximitty-risk-research",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/scenarios/proximitty-risk-research.spec.ts",
            "proofloop/benchmarks/proximitty/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Proximitty has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "proximitty-underwriting-pr0",
          "taskId": "proximitty-underwriting-packet",
          "title": "proximitty-underwriting-packet.spec.ts",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:proximitty:browser",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "proximitty-underwriting-packet",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "proofloop/scenarios/proximitty-underwriting-packet.spec.ts",
            "proofloop/benchmarks/proximitty/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Proximitty has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        }
      ]
    },
    {
      "id": "finch-prod-proxy-task",
      "title": "finch prod proxy task",
      "taskCount": 1,
      "prodLiveBrowserPassed": 1,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 1,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "finch-prod-proxy-task",
          "taskId": "finch-local-financial-evidence-qa",
          "title": "Financial evidence QA through NodeRoom story workflow",
          "status": "prod_live_browser_passed",
          "prodLiveBrowserPassed": true,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run benchmark:proofloop:external-adapter-live-room -- --prod --id finch --real-user --model z-ai/glm-5.2 --model-mode specific",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live"
            }
          },
          "evidence": [
            "proofloop/benchmarks/finch/adapter.json",
            "docs/eval/proofloop-external-adapter-live-room-runs/finch.json",
            "docs/eval/proofloop-proxy-model-sweep.json"
          ],
          "blockers": []
        }
      ]
    },
    {
      "id": "finauditing-prod-proxy-task",
      "title": "finauditing prod proxy task",
      "taskCount": 1,
      "prodLiveBrowserPassed": 1,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 1,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "finauditing-prod-proxy-task",
          "taskId": "finauditing-local-risk-and-misstatement-review",
          "title": "Audit-style evidence review through NodeRoom story workflow",
          "status": "prod_live_browser_passed",
          "prodLiveBrowserPassed": true,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run benchmark:proofloop:external-adapter-live-room -- --prod --id finauditing --real-user --model z-ai/glm-5.2 --model-mode specific",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live"
            }
          },
          "evidence": [
            "proofloop/benchmarks/finauditing/adapter.json",
            "docs/eval/proofloop-external-adapter-live-room-runs/finauditing.json",
            "docs/eval/proofloop-proxy-model-sweep.json"
          ],
          "blockers": []
        }
      ]
    },
    {
      "id": "workstreambench-prod-proxy-task",
      "title": "workstreambench prod proxy task",
      "taskCount": 1,
      "prodLiveBrowserPassed": 1,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 1,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "workstreambench-prod-proxy-task",
          "taskId": "workstreambench-local-spreadsheet-workstream",
          "title": "Spreadsheet workstream execution through NodeRoom story workflow",
          "status": "prod_live_browser_passed",
          "prodLiveBrowserPassed": true,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run benchmark:proofloop:external-adapter-live-room -- --prod --id workstreambench --real-user --model z-ai/glm-5.2 --model-mode specific",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live"
            }
          },
          "evidence": [
            "proofloop/benchmarks/workstreambench/adapter.json",
            "docs/eval/proofloop-external-adapter-live-room-runs/workstreambench.json",
            "docs/eval/proofloop-proxy-model-sweep.json"
          ],
          "blockers": []
        }
      ]
    },
    {
      "id": "noderoom-multi-user-conflict",
      "title": "NodeRoom multi-user conflict suite",
      "taskCount": 6,
      "prodLiveBrowserPassed": 1,
      "localLiveBrowserOnly": 0,
      "runnableProdBrowserTasks": 6,
      "blockedTasks": 0,
      "tasks": [
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-1",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "prod_live_browser_passed",
          "prodLiveBrowserPassed": true,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-1",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts",
            "docs/eval/proofloop-live-multi-user-free-smoke.json",
            "https://noderoom.live/?room=PLMUMR7J1J4G&name=Host",
            "receipt_family=noderoom-multi-user-conflict",
            "receipt_task=multi-user-conflict-1",
            "requested_model=cohere/north-mini-code:free",
            "actual_model=cohere/north-mini-code:free",
            "measured_cost_usd=0"
          ],
          "blockers": []
        },
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-2",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-2",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Multi-user conflict has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-3",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-3",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Multi-user conflict has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-4",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-4",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Multi-user conflict has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-5",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-5",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Multi-user conflict has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        },
        {
          "familyId": "noderoom-multi-user-conflict",
          "taskId": "multi-user-conflict-6",
          "title": "NodeRoom multi-user conflict scenario",
          "status": "ready_for_prod_browser_run",
          "prodLiveBrowserPassed": false,
          "localLiveBrowserOnly": false,
          "runner": {
            "available": true,
            "kind": "playwright_prod_browser",
            "command": "npm run proofloop:live:multi-user-conflict",
            "env": {
              "BENCH_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_BASE_URL": "https://noderoom.live",
              "PLAYWRIGHT_REUSE_SERVER": "1",
              "PROOFLOOP_TASK_IDS": "multi-user-conflict-6",
              "PROOFLOOP_REAL_USER_MODE": "1",
              "PROOFLOOP_FOCUS_MODE": "0",
              "PROOFLOOP_NODEAGENT_RUNTIME_PROFILE": ""
            }
          },
          "evidence": [
            "docs/eval/official-benchmark-task-coverage.json",
            "evals/multiUserCoordinationProof.ts",
            "proofloop/benchmarks/noderoom-multi-user/live-room-scenario.spec.ts",
            "scripts/proofloop-live-playwright.ts"
          ],
          "blockers": [
            "Multi-user conflict has a prod browser adapter, but no passing prod live-browser receipt is recorded for this task/model yet."
          ]
        }
      ]
    }
  ]
}
