pr-review

Coordinate structured GitHub PR reviews with parallel analysis agents.

2|1|Updated Oct 3, 2025
One-click install
npx skills add https://github.com/jamestexas/agents --skill pr-review-jamestexas
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pr-review
Source: https://github.com/jamestexas/agents/tree/main/skills/pr-review
Command: npx skills add https://github.com/jamestexas/agents --skill pr-review-jamestexas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables teams to review and respond to GitHub pull requests with a structured, multi-agent workflow. It orchestrates a preparatory structural analysis, local diagrams, and parallel analysis agents to produce a cohesive review body and inline comments.

Core Features & Use Cases

  • Structural pre-flight via review-prep to surface context and drift
  • Local Mermaid diagrams to visualize PR structure before posting
  • Parallel analysis agents covering convention, behavior, duplication, and infrastructure
  • Synthesis into a single GitHub review with a body and inline comments
  • Respond mode to address reviewer comments

Quick Start

Run /pr-review 12345 to review a PR, or run /pr-review to auto-detect from the current branch.

Frequently Asked Questions about pr-review

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

FAQPage Schema
How do I automate GitHub pull request reviews with multi-agent analysis?

Automated GitHub pull request reviews use parallel analysis agents to evaluate convention, behavior, duplication, and infrastructure. This multi-agent workflow synthesizes structural pre-flight data and local diagrams into a single review body with inline comments.

What is the best way to structure code reviews for behavioral correctness and infrastructure?

Structuring code reviews for behavioral correctness and infrastructure requires applying parallel analysis lenses. These lenses evaluate specific quality dimensions independently before synthesizing the findings into a cohesive GitHub review with targeted inline comments.

Can I visualize pull request structure locally before posting a GitHub review?

Visualizing pull request structure locally before posting a GitHub review is possible by generating local Mermaid diagrams. These diagrams map the PR architecture during the structural pre-flight phase to guide the subsequent parallel analysis.

How do I respond to reviewer comments on a GitHub pull request automatically?

Responding to reviewer comments on a GitHub pull request automatically involves switching to a dedicated Respond mode. This mode processes existing reviewer feedback and coordinates structured replies to address the comments directly.

Does this multi-agent PR review approach work without external dependencies?

This multi-agent PR review approach requires the mache toolkit to fetch pull request data. It relies on this toolkit to extract the necessary context and structural information before applying the parallel analysis agents.

How do I auto-detect a pull request from my current branch for code review?

Auto-detecting a pull request from your current branch for code review requires running the review command without arguments. The system identifies the associated PR and initiates the structural pre-flight and parallel analysis workflow.