Agent skill
receiving-code-review
Use when receiving code review feedback - before implementing suggestions, especially if feedback seems unclear or technically questionable. Requires technical verification and honest evaluation, not performative agreement.
Install this agent skill to your Project
npx add-skill https://github.com/randalmurphal/claude-config/tree/main/skills/receiving-code-review
SKILL.md
Receiving Code Review
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 reacting
2. UNDERSTAND: Restate requirement in own words (or ask)
3. VERIFY: Check against codebase reality
4. EVALUATE: Technically sound for THIS codebase?
5. RESPOND: Technical acknowledgment or reasoned pushback
6. IMPLEMENT: One item at a time, test each
Forbidden Responses
NEVER say:
- "You're absolutely right!"
- "Great point!" / "Excellent feedback!"
- "Let me implement that now" (before verification)
- "Thanks for catching that!"
- Any gratitude expression
INSTEAD:
- Restate the technical requirement
- Ask clarifying questions
- Push back with technical reasoning if wrong
- Just start working (actions > words)
Why no thanks: Actions speak. Just fix it. The code shows you heard the feedback.
Handling Unclear Feedback
IF any item is unclear:
STOP - do not implement anything yet
ASK for clarification on unclear items first
WHY: Items may be related. Partial understanding = wrong implementation.
Example:
Reviewer: "Fix items 1-6"
You understand 1,2,3,6. Unclear on 4,5.
BAD: Implement 1,2,3,6 now, ask about 4,5 later
GOOD: "I understand 1,2,3,6. Need clarification on 4 and 5 before proceeding."
Evaluating External Feedback
Before implementing suggestions from reviewers:
- Technically correct for THIS codebase?
- Breaks existing functionality?
- Reason for current implementation? (it might be intentional)
- Works on all platforms/versions?
- Does reviewer understand full context?
If suggestion seems wrong: Push back with technical reasoning.
If can't verify: Say so: "I can't verify this without [X]. Should I investigate or proceed?"
If conflicts with prior decisions: Stop and discuss with the decision-maker first.
When to Push Back
Push back when:
- Suggestion breaks existing functionality
- Reviewer lacks full context
- Violates YAGNI (unused feature being "properly implemented")
- Technically incorrect for this stack
- Legacy/compatibility reasons exist
- Conflicts with architectural decisions
How to push back:
- Technical reasoning, not defensiveness
- Specific questions
- Reference working tests/code
- Escalate if architectural
YAGNI Check
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
Implementation Order
For multi-item feedback:
- Clarify anything unclear FIRST
- Then implement in this order:
- Blocking issues (breaks, security)
- Simple fixes (typos, imports)
- Complex fixes (refactoring, logic)
- Test each fix individually
- Verify no regressions
Acknowledging Correct Feedback
When feedback IS correct:
GOOD: "Fixed. [Brief description of what changed]"
GOOD: "Good catch - [specific issue]. Fixed in [location]."
GOOD: [Just fix it and move on]
BAD: "You're absolutely right!"
BAD: "Great point!"
BAD: "Thanks for catching that!"
Correcting Your Pushback
If you pushed back and were wrong:
GOOD: "You were right - I checked [X] and it does [Y]. Implementing now."
GOOD: "Verified and you're correct. My understanding was wrong because [reason]. Fixing."
BAD: Long apology
BAD: Defending why you pushed back
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 it breaks things |
| Avoiding pushback | Technical correctness > comfort |
| Partial implementation | Clarify all items first |
GitHub Thread Replies
When replying to inline review comments, reply in the comment thread (gh api repos/{owner}/{repo}/pulls/{pr}/comments/{id}/replies), not as a top-level PR comment.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-prompting
Write effective prompts for Task tool sub-agents, slash commands, and system prompts. Covers Claude 4.x prompting patterns, context engineering, output format specification, and parallel delegation. Use when spawning sub-agents, creating slash commands, or writing system prompts.
test-driven-development
Use when implementing any feature, bug fix, or behavior change - before writing implementation code. Enforces strict RED-GREEN-REFACTOR cycle where tests are written and seen failing before any production code exists.
spec-formats
Templates for brainstorm artifacts and manifest.json. Load when using /spec.
review-initiative
Use after /breakdown-work or manual initiative creation, before running tasks. Use when initiative has 5+ tasks, multiple dependencies, or references a design document.
MCP Integration
Set up and use MCP (Model Context Protocol) servers to extend Claude Code capabilities including PRISM semantic memory, filesystem access, and database integration. Use when setting up MCP servers, debugging MCP connections, or understanding MCP tool usage.
systematic-debugging
Use when debugging any failure, bug, or unexpected behavior - especially before attempting any fix. Enforces root cause investigation before implementation to prevent fix-thrashing and symptom-masking.
Didn't find tool you were looking for?