{
  "schema_version": "theorempath.ask.holdout.seal.v2",
  "sealed_at_utc": "2026-08-10T16:54:15Z",
  "target_origin_main_commit": "dd384574b5f3765acd4a131caae755697a224557",
  "dataset": {
    "path": "holdout_v2.jsonl",
    "sha256": "f70b248bf0f873eeb8912f545d237d55581a1bd5660151091f30424fecee9bf9",
    "bytes": 35175,
    "cases": 96,
    "unique_case_ids": 96,
    "provider_call_labels": 48,
    "no_source_labels": 48,
    "primary_strata": 12,
    "cases_per_primary_stratum": 8
  },
  "protocol": {
    "path": "PROTOCOL_v2.md",
    "sha256": "72a56ac5dd83dfbe324c46244ba3d5c3f201b4dce6fd11a10dad39cd03ab2fe9",
    "bytes": 8071
  },
  "pre_seal_repository_content_inspected": false,
  "pre_seal_prior_holdout_artifacts_inspected": false,
  "pre_seal_tests_or_research_logs_inspected": false,
  "pre_seal_pull_requests_inspected": false,
  "pre_seal_generated_evaluation_outputs_inspected": false,
  "pre_seal_allowed_git_metadata_queries": [
    "git fetch origin main --quiet",
    "git rev-parse origin/main",
    "git status --short --branch"
  ],
  "orchestration_context_disclosure": "The evaluator did not open prior artifacts or exact cases. The orchestration environment automatically supplied a high-level memory summary mentioning aggregate v1 quality figures before task execution. The v2 cases, labels, strata, and thresholds were authored without consulting those figures, repository contents, or prior artifacts. This run is source-blind and exact-case-blind, but not fully history-blind in the strictest sense."
}