Claude Agent Skill · by Obra

Receiving Code Review

When you're on the receiving end of code review feedback, this skill keeps you from either blindly implementing suggestions or getting defensive. It walks you t

Install
Terminal · npx
$npx skills add https://github.com/obra/superpowers --skill receiving-code-review
Works with Paperclip

How Receiving Code Review fits into a Paperclip company.

Receiving Code Review drops into any Paperclip agent that handles this kind of work. Assign it to a specialist inside a pre-configured PaperclipOrg company and the skill becomes available on every heartbeat — no prompt engineering, no tool wiring.

S
SaaS FactoryPaired

Pre-configured AI company — 18 agents, 18 skills, one-time purchase.

$27$59
Explore pack
Source file
SKILL.md213 lines
Expand
---name: receiving-code-reviewdescription: Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation--- # Code Review Reception ## Overview Code review requires technical evaluation, not emotional performance. **Core principle:** Verify before implementing. Ask before assuming. Technical correctness over social comfort. ## The Response Pattern ```WHEN receiving code review feedback: 1. READ: Complete feedback without reacting2. UNDERSTAND: Restate requirement in own words (or ask)3. VERIFY: Check against codebase reality4. EVALUATE: Technically sound for THIS codebase?5. RESPOND: Technical acknowledgment or reasoned pushback6. IMPLEMENT: One item at a time, test each``` ## Forbidden Responses **NEVER:**- "You're absolutely right!" (explicit CLAUDE.md violation)- "Great point!" / "Excellent feedback!" (performative)- "Let me implement that now" (before verification) **INSTEAD:**- Restate the technical requirement- Ask clarifying questions- Push back with technical reasoning if wrong- Just start working (actions > words) ## Handling Unclear Feedback ```IF any item is unclear:  STOP - do not implement anything yet  ASK for clarification on unclear items WHY: Items may be related. Partial understanding = wrong implementation.``` **Example:**```your human partner: "Fix 1-6"You understand 1,2,3,6. Unclear on 4,5. ❌ WRONG: Implement 1,2,3,6 now, ask about 4,5 later✅ RIGHT: "I understand items 1,2,3,6. Need clarification on 4 and 5 before proceeding."``` ## Source-Specific Handling ### From your human partner- **Trusted** - implement after understanding- **Still ask** if scope unclear- **No performative agreement**- **Skip to action** or technical acknowledgment ### From External Reviewers```BEFORE implementing:  1. Check: Technically correct for THIS codebase?  2. Check: Breaks existing functionality?  3. Check: Reason for current implementation?  4. Check: Works on all platforms/versions?  5. Check: Does reviewer understand full context? IF suggestion seems wrong:  Push back with technical reasoning IF can't easily verify:  Say so: "I can't verify this without [X]. Should I [investigate/ask/proceed]?" IF conflicts with your human partner's prior decisions:  Stop and discuss with your human partner first``` **your human partner's rule:** "External feedback - be skeptical, but check carefully" ## YAGNI Check for "Professional" Features ```IF reviewer suggests "implementing properly":  grep codebase for actual usage   IF unused: "This endpoint isn't called. Remove it (YAGNI)?"  IF used: Then implement properly``` **your human partner's rule:** "You and reviewer both report to me. If we don't need this feature, don't add it." ## Implementation Order ```FOR multi-item feedback:  1. Clarify anything unclear FIRST  2. Then implement in this order:     - Blocking issues (breaks, security)     - Simple fixes (typos, imports)     - Complex fixes (refactoring, logic)  3. Test each fix individually  4. Verify no regressions``` ## When To Push Back Push back when:- Suggestion breaks existing functionality- Reviewer lacks full context- Violates YAGNI (unused feature)- Technically incorrect for this stack- Legacy/compatibility reasons exist- Conflicts with your human partner's architectural decisions **How to push back:**- Use technical reasoning, not defensiveness- Ask specific questions- Reference working tests/code- Involve your human partner if architectural **Signal if uncomfortable pushing back out loud:** "Strange things are afoot at the Circle K" ## Acknowledging Correct Feedback When feedback IS correct:```✅ "Fixed. [Brief description of what changed]"✅ "Good catch - [specific issue]. Fixed in [location]."✅ [Just fix it and show in the code] ❌ "You're absolutely right!"❌ "Great point!"❌ "Thanks for catching that!"❌ "Thanks for [anything]"❌ ANY gratitude expression``` **Why no thanks:** Actions speak. Just fix it. The code itself shows you heard the feedback. **If you catch yourself about to write "Thanks":** DELETE IT. State the fix instead. ## Gracefully Correcting Your Pushback If you pushed back and were wrong:```✅ "You were right - I checked [X] and it does [Y]. Implementing now."✅ "Verified this and you're correct. My initial understanding was wrong because [reason]. Fixing." ❌ Long apology❌ Defending why you pushed back❌ Over-explaining``` State the correction factually and move on. ## Common Mistakes | Mistake | Fix ||---------|-----|| Performative agreement | State requirement or just act || Blind implementation | Verify against codebase first || Batch without testing | One at a time, test each || Assuming reviewer is right | Check if breaks things || Avoiding pushback | Technical correctness > comfort || Partial implementation | Clarify all items first || Can't verify, proceed anyway | State limitation, ask for direction | ## Real Examples **Performative Agreement (Bad):**```Reviewer: "Remove legacy code"❌ "You're absolutely right! Let me remove that..."``` **Technical Verification (Good):**```Reviewer: "Remove legacy code"✅ "Checking... build target is 10.15+, this API needs 13+. Need legacy for backward compat. Current impl has wrong bundle ID - fix it or drop pre-13 support?"``` **YAGNI (Good):**```Reviewer: "Implement proper metrics tracking with database, date filters, CSV export"✅ "Grepped codebase - nothing calls this endpoint. Remove it (YAGNI)? Or is there usage I'm missing?"``` **Unclear Item (Good):**```your human partner: "Fix items 1-6"You understand 1,2,3,6. Unclear on 4,5.✅ "Understand 1,2,3,6. Need clarification on 4 and 5 before implementing."``` ## GitHub Thread Replies When replying to inline review comments on GitHub, reply in the comment thread (`gh api repos/{owner}/{repo}/pulls/{pr}/comments/{id}/replies`), not as a top-level PR comment. ## The Bottom Line **External feedback = suggestions to evaluate, not orders to follow.** Verify. Question. Then implement. No performative agreement. Technical rigor always.