{
  "model": "claude-sonnet-5-5",
  "run_date": "2026-10-01",
  "tasks": 12,
  "k": 2,
  "A": {
    "passes": 4,
    "trials": 24,
    "wilson_ci95": [
      0.07,
      0.36
    ],
    "checks_passed": 165,
    "checks_total": 242
  },
  "S": {
    "passes": 14,
    "trials": 24,
    "wilson_ci95": [
      0.39,
      0.76
    ],
    "checks_passed": 217,
    "checks_total": 242
  },
  "paired_delta": 0.42,
  "task_bootstrap_ci95": [
    0.08,
    0.71
  ],
  "exposure": {
    "loaded": 24,
    "total": 24
  },
  "privacy": "Check diagnostic values, account labels, fixtures and transcripts omitted."
}