aif-review-project-contextlisted

# aif-review skill-context — anti-tautology two-AI review conventions > **Authoritative for:** project-specific review conventions injected into AI Factory's `aif-review` skill (and its background `review-sidecar`) — the anti-tautology / two-AI test-review checks this project requires in addition to AIF's generic review. > **NOT authoritative for:** project goal — see the consumer's README.md. The generic correctness/security/performance review — that stays owned by AIF's `aif-review` SKILL.md; this file augments it. <!-- @dual-pair: review-sidecar --> <!-- spec-of: agents/review-sidecar.md --> <!-- Portable SSOT for this content is agents/review-sidecar.md. This file is the AIF-native delivery channel (skill-context override) for the same anti-tautology spec, per .claude/rules/dual-implementation-discipline.md §7. Edit the SSOT first, then mirror here. --> This project has learned, from repeated incidents, that the highest-value review signal is **test quality**, not just code correctness. AIF's generic review already covers correctness / security / performance; the conventions below are what this project additionally requires. Apply them as project-level overrides — when they add a check AIF's defaults don't have, perform that check and include it in the output. ## Review the diff as a cold external reviewer Read the changed code as if you had never seen it and did **not** write it. The implementer wrote the code and its tests in one head — same blind spots.