skills.rest
cfactolerin
cfactolerinCOMMUNITY
Explore all skills by this author →
GitHub

code-reviewer-start

Run multi-agent code reviews on local branches before pushing to GitHub.

Updated May 18, 2026
Search Tags:#code review#Claude#Codex#GitHub#Gemini#opencode
One-click install
npx skills add https://github.com/cfactolerin/code_reviewer --skill code-reviewer-start
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-reviewer-start
Source: https://github.com/cfactolerin/code_reviewer/tree/main/skills/code-reviewer-start
Command: npx skills add https://github.com/cfactolerin/code_reviewer --skill code-reviewer-start

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill performs a non-interactive, multi-agent code review on the current local branch before pushing to GitHub, catching bugs, security issues, regressions, and missing tests automatically.

Core Features & Use Cases

  • Parallel Code Review: Simultaneously reviews the branch diff against the base branch using Claude, Codex, Gemini, and opencode.
  • Automated Arbiter: A Claude-powered arbiter cross-examines disagreements and produces a single FINAL_REVIEW_RESULTS.md.
  • Customizable Review: Supports flags like --delta for reviewing only new commits/changes since last review and --full for a complete branch review.

Quick Start

Run /code-reviewer:start to start the review process.

Frequently Asked Questions about code-reviewer-start

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

FAQPage Schema
How do I run an automated code review before pushing to GitHub?▼

To run an automated code review before pushing to GitHub, trigger the process to analyze your local branch diff. It simultaneously uses multiple AI agents to catch bugs, security issues, regressions, and missing tests before code reaches the remote repository.

Can I use multiple AI agents like Claude, Codex, and Gemini together for code review?▼

Yes, you can use Claude, Codex, Gemini, and opencode together for code review. The process runs these agents in parallel to analyze branch diffs, with a Claude-powered arbiter cross-examining disagreements to produce final unified results.

What is the difference between a delta review and a full branch review?▼

A delta review analyzes only new commits and changes since the last review, while a full branch review evaluates the complete branch diff against the base branch. Both methods help catch bugs and security issues before pushing to GitHub.

How does a multi-agent code review resolve disagreements between AI models?▼

Multi-agent code review resolves disagreements using a Claude-powered arbiter. This arbiter cross-examines conflicting findings from Codex, Gemini, and opencode, consolidating the feedback into a single FINAL_REVIEW_RESULTS.md file.

Do I need a Jira ticket key to run a local code review on my branch?▼

No, a Jira ticket key is not strictly required to run a local code review. The process supports configurable flags for Jira ticket keys and base branches, allowing you to customize the review context without mandatory external tracking dependencies.

Related Skills

Explore More Software Engineering Skills→
smallnestsmallnest
207

review-it

Review local or branch-based code changes before commit or ship.

Community
Intermediate
OpenAuthorityOpenAuthority
8

git_push

Push local branch commits to a remote Git repository.

Official
Intermediate
gooftroopgooftroop

mastering-github-cli

Search GitHub repositories, monitor CI/CD pipelines, and manage Actions workflows via CLI.

Community
Advanced
onairmarconairmarc
2

code-review

Compare the current branch against main and report breaking changes, code quality issues, test gaps, and framework-specific patterns.

Community
Advanced