skills.rest
heyu-ai
heyu-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

pr-cycle-deep

Automate PR reviews with multi-agent mob review and cross-model debate.

Updated May 15, 2026
Search Tags:#code review#PR review#GitHub integration#mob review#cross-model debate
One-click install
npx skills add https://github.com/heyu-ai/yibi-stack --skill pr-cycle-deep
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pr-cycle-deep
Source: https://github.com/heyu-ai/yibi-stack/tree/main/plugins/pr-flow/skills/pr-cycle-deep
Command: npx skills add https://github.com/heyu-ai/yibi-stack --skill pr-cycle-deep

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires codex, gemini, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates and enhances PR review processes by enabling a multi-agent mob review with cross-model debate, significantly reducing manual effort and improving the quality of reviews.

Core Features & Use Cases

  • Multi-Agent Mob Review: Leverages Claude, Codex, and Gemini for independent reviews, fostering a more comprehensive understanding of the code changes.
  • Cross-Model Debate: Allows each agent to read and respond to others' findings, ensuring a robust debate and aggregated final report.
  • Use Case: Ideal for medium to large PRs, high-risk changes, or projects requiring cross-family LLM review to ensure consistency in flagged issues.

Quick Start

Initiate a PR review cycle with the command '/pr-cycle-deep'.

Frequently Asked Questions about pr-cycle-deep

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

FAQPage Schema
How do I automate GitHub PR review with multi-agent debate?▼

To automate GitHub PR review with multi-agent debate, this Skill leverages Claude, Codex, and Gemini to independently review changes and aggregate findings into a final JSON report via GitHub API.

What is cross-model debate in code review?▼

Cross-model debate in code review is a process where multiple AI agents read and respond to each other's findings, ensuring a robust discussion that flags high-risk changes and improves consistency before generating an aggregated report.

Do I need Codex and Gemini to run a mob review for high-risk PRs?▼

Yes, you need both Codex and Gemini dependencies to run a mob review. They provide the cross-family LLM review capabilities required to independently analyze medium to large PRs and debate code changes.

How do I start a PR review cycle for large pull requests?▼

You can start a PR review cycle for large pull requests by running the '/pr-cycle-deep' command, which triggers the multi-agent review and cross-model debate process using GitHub API operations.

Are there limitations to using multi-agent code review for GitHub PRs?▼

Limitations of this multi-agent code review include the strict dependency on both Codex and Gemini for functionality, and it is specifically designed for medium to large PRs rather than small, trivial changes.

What is the best way to review high-risk code changes across different LLMs?▼

The best way to review high-risk code changes across different LLMs is using a mob review approach where Claude, Codex, and Gemini independently evaluate the PR and debate findings to ensure comprehensive issue detection.

Related Skills

Explore More Software Engineering Skills→
gqf2008gqf2008

github-auth

Authenticate Hermes Agent with GitHub via gh CLI or git tokens and SSH keys.

Community
Intermediate
j0rugej0ruge
8

coderabbit_pr

Parse GitHub PR AI review comments and generate per-reviewer markdown checklists.

Community
Advanced
KrismithReddy12KrismithReddy12

health

Calculate a weighted 0-10 code quality score from linters and test runners.

Community
Advanced
heyu-aiheyu-ai

mob-code-review-only

Automate external PR code reviews using multi-model collaboration.

Official
Advanced