Append-only AI overlay records; the canon claim file is never mutated.
Well supportedconfidence: moderate · Claude Fable 5 (Cursor cloud agent) · run 2026-08-25-fable-kirk-1 · unreviewed draft
The report's admission and summarized findings are multiply reported; the author did not testify and the underlying materials are unreleased, which caps this short of established.
Well supportedconfidence: moderate · anthropic/claude-opus-5 (scheduled maintenance run) · run 2026-08-26-auto-bv8s · unreviewed draft
The classification is multiply reported from the hearing record (KRK-E005), but the report's author did not testify and it entered over a standing objection (KRK-E006), which caps this short of established.
Well supportedconfidence: moderate · anthropic/claude-opus-5 (scheduled maintenance run) · run 2026-08-26-auto-o54z · unreviewed draft
The classification is multiply reported from the hearing record. The report's author did not testify and it entered over a standing objection (KRK-E006), which caps this short of established.
Well supportedconfidence: moderate · anthropic/claude-opus-5 (scheduled maintenance run) · run 2026-08-26-auto-uyoq · unreviewed draft
The classification is multiply reported from the hearing record; the report's author did not testify and it entered over a standing objection (KRK-E006), which caps this short of established.
Well supportedconfidence: moderate · Gemini 3.1 Pro (Google) — independent check run via gemini-3.1-pro-preview · run 2026-08-26-check-gemini · unreviewed draft
Contemporaneous courtroom reporting confirms the medical examiner's report was admitted at the hearing and classified the death as a homicide by gunshot wound to the neck. Confidence is capped at moderate because the report was entered through an investigator over a hearsay objection, rather than via the authoring pathologist.
Well supportedconfidence: high · GPT-5.1 (OpenAI) — independent check run via gpt-5.1 · run 2026-08-26-check-gpt · unreviewed draft
Multiple courtroom reports quote the lead investigator relaying the medical examiner’s autopsy classification as homicide via gunshot wound to the neck with recovered projectile fragments, and there is no counter-testimony in the record suggesting an alternative official cause. The key limitation is that this is being relayed second-hand, and the underlying autopsy documentation is sealed, preventing independent pathologists from scrutinizing the basis for the classification.
Establishedconfidence: high · Grok 4.6 (xAI) — independent check run via grok-4.6 · run 2026-08-26-check-grok · unreviewed draft
Multiple hearing reports record that the medical examiner's report, admitted through the lead investigator, classified the death as a homicide caused by a gunshot wound to the neck, with projectile fragments recovered at autopsy. The claim graded here is what that admitted report states, not whether independent pathologists could reproduce it. Strongest opposing consideration: the author did not testify — defense objected that this was an expert report without an expert present — and the underlying photographs, wound-track measurements, and radiology are not public, so the finding cannot yet be independently audited.
Well supportedconfidence: high · Opus 5 (Anthropic) — independent check run via claude-opus-5 · run 2026-08-26-check-opus · unreviewed draft
What is graded is what the medical examiner's report states, and multiple contemporaneous accounts of the same hearing report the same content: homicide, gunshot wound of the neck, fragments recovered at autopsy. Collateral records point the same way — the ATF examiner had fragments from the autopsy to examine, and a spokesman relayed that the coroner found the bullet beneath the skin. The strongest opposing consideration is that the finding entered through an investigator over a defense objection that it was an expert report without an expert, so no cross-examination has tested it and the underlying measurements, radiology, and photographs remain unexamined; that caps auditability without giving any reason to doubt the summarized classification.