receiving-code-review

The receiving-code-review skill acts as a critical evaluation layer for AI coding agents when processing feedback from human partners or external reviewers. It shifts the agent's behavior from passive acceptance to active technical validation, ensuring that every suggestion is cross-referenced with codebase reality, performance constraints, and architectural standards before implementation. This skill is designed for autonomous or semi-autonomous development environments where technical correctness, security, and the YAGNI (You Aren't Gonna Need It) principle are prioritized over social comfort or conversational pleasantries.

• Systematic feedback processing: Guides the agent through a six-step pipeline including reading, restating requirements, verifying against codebase, evaluating technical soundness, responding, and granular task implementation.

• Anti-performative response patterns: Explicitly forbids empty, polite, or performative agreements like 'You're right' or 'Great point', requiring instead technical acknowledgments or reasoned pushbacks based on code analysis.

• Automated verification logic: Enforces rigorous checks on proposed changes, requiring the agent to verify if suggestions break existing functionality, violate project context, or contradict established human-partner decisions.

• Intelligent pushback workflows: Provides clear paths for technical disagreement, requiring the agent to cite build targets, legacy dependencies, or architectural conflicts when a suggestion appears incorrect or misguided.

• Granular execution strategies: Mandates one-item-at-a-time implementation with individual testing to prevent regressions and batch-processing errors.

• Always verify technical claims by grepping the codebase and checking for usage patterns before accepting external suggestions.

• Use this skill when receiving feedback to manage multi-item lists, ensuring all items are understood before any code is modified.

• If a feedback item is unclear, stop immediately and ask for clarification rather than implementing a 'best guess' solution.

• Always prioritize technical reasoning and existing test results over the perceived social pressure of the review process.

• Use for any task involving code changes, PR comments, or direct suggestions from human developers or automated tools.