Reviews¶
Each review round writes one file here, <round-id>.md (e.g. R1-consistency.md), produced by the /review-round skill. Findings are applied to the specs only after triage, by /apply-findings.
Finding format¶
### SR-012 · major · Gate decisions: two definitions of "exclude"
- **Where:** api-and-data-model.md §5 (gate decision body) vs checking-scenarios.md §5 (decision types)
- **Evidence:** "…" (api §5) / "…" (checking §5)
- **Problem:** an implementer cannot tell whether excluding an item also edits the samplesheet.
- **Proposal:** define `exclude` once in checking-scenarios.md §5; api §5 links to it and lists the fields.
- **Found by:** spec-reviewer, constraints-hunter
- **Status:** open | accepted | rejected (reason) | applied (commit / section)
Severity:
- Blocker: implementation cannot start, or would build the wrong thing.
- Major: a real problem that will cost rework.
- Minor: worth fixing, low risk.
ID prefixes:
SR: spec reviewer;FC: fact checker;CH: constraints hunter;SEC: security;UX: UX reviewer;DL: depictio liaison;TA: test architect;IE: infra engineer;DEP-T: depictio liaison, tooling and ecosystem re-audit (R8b).