receiving-code-review

Guides AI agents to verify, evaluate, and implement code review feedback with technical reasoning.

Updated Jan 2, 2025
One-click install
npx skills add https://github.com/mcinnisd/gymbro --skill receiving-code-review-mcinnisd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: receiving-code-review
Source: https://github.com/mcinnisd/gymbro/tree/main/.agents/skills/receiving-code-review
Command: npx skills add https://github.com/mcinnisd/gymbro --skill receiving-code-review-mcinnisd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents often respond to code review feedback with performative agreement or blind implementation, leading to unverified changes, broken functionality, and wasted effort. This Skill enforces a disciplined response pattern: verify before implementing, ask before assuming, and prioritize technical correctness over social comfort. ## Core Features & Use Cases - Structured Response Pattern: Read, understand, verify, evaluate, respond, and implement feedback one item at a time with individual testing. - Source-Specific Handling: Differentiates trusted human partner feedback from external reviewer suggestions, applying skepticism and codebase verification to the latter. - Pushback and YAGNI Checks: Provides rules for pushing back on incorrect suggestions, checking for unused features before implementing, and gracefully correcting wrong pushback. - Use Case: When a reviewer on a GitHub pull request suggests removing legacy code or adding a metrics endpoint, the agent first greps the codebase for actual usage, verifies platform compatibility, and replies in the review thread with technical reasoning instead of blind agreement. ## Quick Start Apply the receiving-code-review skill to evaluate and respond to the review comments on my current pull request.

Frequently Asked Questions about receiving-code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How should an AI agent respond to code review feedback?

An AI agent should read the full feedback, restate the requirement, verify it against the codebase, evaluate technical soundness, then respond with acknowledgment or reasoned pushback. It should implement one item at a time and test each fix individually.

How do I handle unclear code review comments before implementing?

Stop and ask for clarification on any unclear items before implementing anything. Partial understanding leads to wrong implementations because feedback items are often related, so clarify all items first rather than batching the clear ones.

When should you push back on a code review suggestion?

Push back when a suggestion breaks existing functionality, violates YAGNI for unused features, is technically incorrect for the stack, or conflicts with prior architectural decisions. Use technical reasoning and reference working tests or code rather than being defensive.

Should AI agents say thank you for code review feedback?

No. The skill forbids gratitude expressions and performative agreement like "Great point!" or "You're absolutely right!" Instead, state the fix briefly or just implement it, letting the code itself demonstrate the feedback was heard.

How do I reply to inline GitHub review comments?

Reply within the comment thread using the GitHub API endpoint for pull request comment replies, not as a top-level PR comment. This keeps the discussion attached to the specific line of code being reviewed.

What is the YAGNI check for reviewer-suggested features?

Before implementing a suggested feature, grep the codebase for actual usage of the relevant endpoint or function. If nothing calls it, propose removing it under YAGNI rather than building out an unused capability.