co-review

Reconcile feedback from multiple local AI agents and GitHub comments into a single review.

3|Updated May 7, 2026
One-click install
npx skills add https://github.com/bestdan/workflow-skills --skill co-review-bestdan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: co-review
Source: https://github.com/bestdan/workflow-skills/tree/main/skills/co-review
Command: npx skills add https://github.com/bestdan/workflow-skills --skill co-review-bestdan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, git, and includes scripts (resource) components.

What problem does it solve?

This Skill solves the problem of fragmented and biased code reviews by reconciling feedback from multiple local AI agents and GitHub comments into a single, actionable, and auditable review.

Core Features & Use Cases

  • Multi-Agent Reconciliation: Aggregates reviews from local agents like Codex, Devin, and Copilot, using a sub-agent to independently verify findings.
  • Automated Fixes: Automatically applies high-confidence fixes to your working tree or posts vetted findings back to GitHub.
  • Use Case: When working on a complex feature branch, use this to run a comprehensive review that combines your own analysis with feedback from multiple specialized coding agents, ensuring all edge cases are covered before merging.

Quick Start

Run a collaborative review on your current pull request by typing /co-review in your terminal.

Frequently Asked Questions about co-review

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

FAQPage Schema
How do I automate code review using multiple AI agents?

Yes, automated PR review can apply high-confidence fixes directly to your working tree. The Skill evaluates reconciled feedback from multiple agents and automatically patches the code or posts vetted findings back to GitHub comments.

Do I need the GitHub CLI to reconcile pull request review comments?

Yes, you need the GitHub CLI and git installed to retrieve diffs and existing GitHub comments. The Skill depends on these CLI tools to fetch pull request data and post the reconciled review feedback back to the repository.

What is multi-agent code review reconciliation?

Multi-agent code review reconciliation is the process of aggregating feedback from several local AI coding agents and existing GitHub comments. It uses an independent sub-agent to verify findings, ensuring a comprehensive and auditable review that eliminates fragmented analysis.

Can I run a collaborative code review on a local working tree instead of a pull request?

Yes, you can run a collaborative code review on a local working tree. The Skill operates directly on local changes to identify bugs and enforce project conventions, allowing you to catch edge cases before creating a pull request.

What are the limitations of automated code refactoring from AI agent reviews?

The limitation of automated code refactoring here is that only high-confidence fixes are applied automatically. Lower confidence findings and complex architectural changes are posted as vetted findings back to GitHub for manual human review instead of direct patching.