{
  "runDate": "2026-09-16",
  "modelLabel": "GPT-6 Astra",
  "taskCount": 10,
  "sampleSize": 40,
  "configurationCount": 2,
  "repetitionsPerTaskSetting": 2,
  "outcomes": {
    "passed": 30,
    "failed": 9,
    "invalid": 1
  },
  "provenance": {
    "sourceBoundarySha256": "ef20f0839cea66d43a046b3308de366e35270c5d9e91ce1beba91d0d4ddf68c0",
    "methods": "/studies/api-patch/methods.json",
    "methodsVersion": "2026-09-27.1",
    "verification": "source and archived receipts matched; no execution rerun or human validation"
  },
  "taskResults": [
    {
      "label": "Project priority filter",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 4
    },
    {
      "label": "Python pagination authentication",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 2
    },
    {
      "label": "CLI cursor pagination",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 4
    },
    {
      "label": "Bounded rate-limit retries",
      "outcomes": {
        "passed": 3,
        "failed": 0,
        "invalid": 1
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 3
    },
    {
      "label": "Backwards-compatible CLI option",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 4
    },
    {
      "label": "Nullable completion times",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 4
    },
    {
      "label": "JSON-mode process behavior",
      "outcomes": {
        "passed": 4,
        "failed": 0,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 0,
      "passedWithTestCheckFailure": 0
    },
    {
      "label": "Idempotent job creation",
      "outcomes": {
        "passed": 0,
        "failed": 4,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 4,
      "passedWithTestCheckFailure": 0
    },
    {
      "label": "Backwards-compatible job status",
      "outcomes": {
        "passed": 3,
        "failed": 1,
        "invalid": 0
      },
      "functionalFailures": 0,
      "documentationFailures": 1,
      "passedWithTestCheckFailure": 0
    },
    {
      "label": "Wait for job completion",
      "outcomes": {
        "passed": 0,
        "failed": 4,
        "invalid": 0
      },
      "functionalFailures": 4,
      "documentationFailures": 3,
      "passedWithTestCheckFailure": 0
    }
  ],
  "optionalCheck": {
    "byOverallOutcome": {
      "passed": {
        "passed": 9,
        "failed": 21,
        "ungraded": 0
      },
      "failed": {
        "passed": 0,
        "failed": 9,
        "ungraded": 0
      },
      "invalid": {
        "passed": 0,
        "failed": 0,
        "ungraded": 1
      }
    },
    "missingOrSkippedGradeableChecks": 0,
    "verification": "matched archived receipts; no execution rerun or human validation"
  }
}