Can I add my own review criteria?

Yes, the skill is highly extensible. You can create new review skills in the .claude/skills/collaboration/reviews/ directory to add custom analysis dimensions.

What file types are supported?

It currently supports AI prompt content, Python files, and general Pull Request file changes, with the architecture to support any file type via new sub-skills.

Does this skill perform the review itself?

No, it acts as an orchestrator that delegates the actual review tasks to specialized sub-skills to ensure expert-level analysis for each specific domain.

Why use subagents instead of a single review?

Using specialized subagents allows for deeper, more focused analysis on specific patterns like token efficiency or type safety, leading to significantly higher quality feedback than a general-purpose review.

Does it support parallel execution?

Yes, it is designed to launch all necessary review subagents in parallel to minimize wait times and maximize processing efficiency.

Code Review Orchestrator

Name: Code Review Orchestrator
Author: jack-michaud

byjack-michaud

•

协作工具

Streamlines code reviews by delegating specialized analysis to subagents and synthesizing findings into a unified report.

The Code Review Orchestrator is a specialized Claude Code skill designed to manage complex review workflows by coordinating multiple specialized subagents. Instead of performing a generic review, it identifies the appropriate review dimensions—such as prompt brevity, Python type safety, or PR artifact cleanup—and launches parallel agents to provide focused, high-quality feedback. This ensures that every line of code or documentation is assessed against specific best practices without manual intervention, making it an essential tool for maintaining high standards in AI-assisted development and large-scale repositories.

主要功能

01Support for reviewing AI prompts, Python code, and Pull Request artifacts

02Extensible architecture allowing users to add custom review dimensions easily

03Parallel execution of specialized review subagents for maximum efficiency

04Automated mapping of file types to relevant domain-specific review skills

05Standardized reporting with line-specific suggestions and technical reasoning

063 GitHub stars

使用场景

01Auditing AI prompt efficiency and token usage within Claude skills

02Enforcing Python type-hinting standards across large codebases

03Automating multi-dimensional code reviews during the Pull Request process

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jack-michaud/faire code-review

For use in Claude.ai and ChatGPT

Download Skill