← ClaudeAtlas

receiving-code-reviewlisted

Use when a Phase B agent receives bot review findings — before implementing any suggestion, especially when a finding conflicts with sibling patterns in the codebase or contradicts another reviewer. Judgment layer only; the mechanical fix/reply/resolve loop lives in cr-github-review.md.
auerbachb/claude-code-config · ★ 5 · Code & Development · score 73
Install: claude install-skill auerbachb/claude-code-config
<!-- Adapted from obra/superpowers skills/receiving-code-review/SKILL.md @ b36e0829 --> # Receiving Code Review — Judgment Layer Code review requires technical evaluation, not emotional performance. This skill guides the EVALUATE step that sits between reading a bot finding and touching any code. **Core principle:** Verify before implementing. Technical correctness over performative compliance. > **Scope note:** This file is the *judgment* layer. The mechanical loop — polling endpoints, batching fixes, replying to threads, resolving via GraphQL — belongs to `cr-github-review.md`. Do not duplicate that loop here. ## The Six-Step Response Pattern ``` WHEN a bot reviewer (coderabbitai[bot], cursor[bot], greptile-apps[bot], codeant-ai[bot], graphite-app[bot]) posts a finding: 1. READ: Complete the finding without reacting 2. UNDERSTAND: Restate the requirement in your own words (or flag if unclear) 3. VERIFY: Check against codebase reality — open the actual file 4. EVALUATE: Is this finding technically sound FOR THIS codebase? 5. RESPOND: Technical acknowledgment or reasoned decline with evidence 6. IMPLEMENT: Only valid findings, one at a time, per cr-github-review.md ``` **STOP before step 6** if you have not completed steps 3 and 4. Incomplete verification = incomplete review. ### Step 3 — VERIFY maps to our verification step "Check against the actual file" in step 3 is the same verification step described in `cr-local-review.md`. Specifica