review-pr

Run parallel local and remote pull request reviews and synthesize deduplicated findings.

2|Updated Sep 3, 2016
One-click install
npx skills add https://github.com/jpaddison3/dotfiles --skill review-pr-jpaddison3
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-pr
Source: https://github.com/jpaddison3/dotfiles/tree/main/claude-skills/review-pr
Command: npx skills add https://github.com/jpaddison3/dotfiles --skill review-pr-jpaddison3

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates and consolidates multiple review sources—local AI agents, Codex, and remote GitHub reviewers—so maintainers get a single deduplicated, prioritized list of issues instead of sifting through fragmented reviewer outputs.

Core Features & Use Cases

  • Launches parallel local AI review agents (multiple Claude subagents and Codex) and a remote review fetch to gather Gemini, Copilot, and human comments.
  • Supports modes for full, mini (fast/cheap), passthrough, and autonomous fix flows, plus base-branch overrides and PR URL detection.
  • Evaluates remote bot readiness, deduplicates findings into canonical issues, assigns IDs, and formats collaborative or passthrough reports suitable for merge decisions.
  • Use case: on a feature branch, run a single command to run local reviews, fetch remote bot/human feedback, and receive a synthesized, prioritized review summary with remediation guidance.

Quick Start

Run /review-pr to auto-detect the current branch PR, run local and remote reviews in parallel, and return a synthesized collaborative review.

Frequently Asked Questions about review-pr

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

FAQPage Schema
How do I consolidate AI and human pull request reviews into a single deduplicated list?

To consolidate pull request reviews, this Skill runs parallel local AI agents and remote GitHub fetches, deduplicates findings into canonical issues, and emits a synthesized, prioritized review summary for merge decisions.

Can I run Claude and Codex code reviews in parallel with remote GitHub bot feedback?

Yes, you can run local Claude subagents and Codex in parallel with remote GitHub bot fetches, correlating and deduplicating outputs from Gemini, Copilot, and humans into a single collaborative report.

What is the best way to automate pull request review synthesis for multiple AI agents?

Automating pull request review synthesis involves launching parallel local and remote review processes, evaluating bot readiness, assigning canonical IDs to deduplicated issues, and formatting collaborative or passthrough reports.

Does this PR review tool support fast mini modes and autonomous fix flows?

Yes, this PR review tool supports full, mini (fast/cheap), passthrough, and autonomous fix flow modes, plus base-branch overrides and automatic PR URL detection to streamline the review process.

How do I deduplicate overlapping code review findings from different automated reviewers?

Deduplicating overlapping code review findings requires evaluating remote bot readiness, correlating outputs from local and remote agents, and assigning canonical numbers to synthesized issues for clear tracking.