When is the best time to request a code review?

Reviews are mandatory after completing major features or before merging to main, and highly recommended after each individual task in a subagent-driven workflow.

What should I do if the AI reviewer provides incorrect feedback?

The workflow encourages pushing back with technical reasoning or providing specific code and tests that prove the implementation is correct.

Does this skill require specific git configurations?

It relies on standard git SHAs (Base and Head) to define the scope of the review, making it compatible with most standard git-based development environments.

How does this skill preserve my session context?

It dispatches a separate subagent with only the necessary git SHAs and requirements, ensuring the reviewer focuses on the code changes rather than your chat history.

Engineering Code Reviewer

Name: Engineering Code Reviewer
Author: Moonsong-Labs

byMoonsong-Labs

•

Collaboration Tools

Automates the dispatch of specialized AI subagents to perform focused code reviews based on git diffs and implementation requirements.

The Requesting Code Review skill streamlines the quality assurance process by dispatching a dedicated code-reviewer subagent to evaluate code changes before they are merged. By providing specific git SHAs and implementation requirements, it ensures the reviewer focuses exclusively on the work product rather than the session history, preserving the developer's context while catching critical logic errors, architecture issues, and missing requirements early in the development cycle. It is an essential tool for subagent-driven development, promoting a 'review early, review often' philosophy to prevent issue cascading.

Key Features

01Context-isolated subagent dispatch for objective code evaluation

02SHA-based diff analysis for precise review scope

03Standardized feedback categorization (Critical, Important, Minor)

04Seamless integration with multi-step development workflows

053 GitHub stars

06Requirement validation against original project plans

Use Cases

01Performing final quality audits before merging features into the main branch

02Verifying individual task completion during subagent-driven development

03Gaining a fresh technical perspective when stuck on complex refactoring or bug fixes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add moonsong-labs/knowledge-work-plugins requesting-code-review

For use in Claude.ai and ChatGPT

Key Features

01Context-isolated subagent dispatch for objective code evaluation

02SHA-based diff analysis for precise review scope

03Standardized feedback categorization (Critical, Important, Minor)

04Seamless integration with multi-step development workflows

053 GitHub stars

06Requirement validation against original project plans

Use Cases

01Performing final quality audits before merging features into the main branch

02Verifying individual task completion during subagent-driven development

03Gaining a fresh technical perspective when stuck on complex refactoring or bug fixes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add moonsong-labs/knowledge-work-plugins requesting-code-review

For use in Claude.ai and ChatGPT