Why does the reviewer subagent not see my session history?

This isolation ensures the reviewer focuses purely on the code product and requirements, preventing the AI's internal thought process from biasing the technical evaluation.

When is it mandatory to use this skill?

It should be used after each task in subagent-driven development, after completing major features, and always before merging into the main branch.

How does the skill identify which code to review?

The skill utilizes git SHAs (Base and Head) to generate a precise diff of the changes made during the specific task or session.

What should I do if I disagree with the AI reviewer's feedback?

You should push back with technical reasoning or provide tests that prove the code works; the skill is designed to facilitate technical dialogue.

Requesting Code Review

Name: Requesting Code Review
Author: zhengxuyu

byzhengxuyu

0•

Collaboration Tools

Automates the process of dispatching specialized subagents to perform focused, context-aware code reviews using git history and specific task requirements.

This skill streamlines the development lifecycle by enabling Claude to initiate rigorous code reviews via a dedicated subagent. By leveraging git SHAs to isolate changes and providing specific implementation plans, it ensures that every feature or bug fix is evaluated against its original requirements without the noise of the current conversation history. This promotes higher code quality, catches potential bugs early in the subagent-driven development flow, and facilitates smoother merges into main branches by categorizing feedback into prioritized action items.

Key Features

01Structured Feedback Classification: Categorizes findings into Critical, Important, and Minor issues for prioritized fixing.

02Requirement Validation: Cross-references implemented code against specific plans and documentation placeholders.

03Isolated Subagent Review: Dispatches a focused reviewer subagent without exposing session history to prevent bias.

040 GitHub stars

05Git-Based Delta Analysis: Uses git SHAs to pinpoint exactly which lines of code need evaluation for precise feedback.

06Subagent-Driven Development Integration: Optimized for use after every task to prevent the cascade of technical issues.

Use Cases

01Performing a final technical audit of a major feature branch before merging to main.

02Verifying individual task completion in subagent-driven development cycles.

03Gaining a fresh perspective and baseline check before starting a complex refactoring project.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zhengxuyu/claude-superpowers requesting-code-review

For use in Claude.ai and ChatGPT

Key Features

01Structured Feedback Classification: Categorizes findings into Critical, Important, and Minor issues for prioritized fixing.

02Requirement Validation: Cross-references implemented code against specific plans and documentation placeholders.

03Isolated Subagent Review: Dispatches a focused reviewer subagent without exposing session history to prevent bias.

040 GitHub stars

05Git-Based Delta Analysis: Uses git SHAs to pinpoint exactly which lines of code need evaluation for precise feedback.

06Subagent-Driven Development Integration: Optimized for use after every task to prevent the cascade of technical issues.

Use Cases

01Performing a final technical audit of a major feature branch before merging to main.

02Verifying individual task completion in subagent-driven development cycles.

03Gaining a fresh perspective and baseline check before starting a complex refactoring project.