Judge

Automate code reviews with codex review CLI to detect bugs and security vulnerabilities.

68|14|Updated Jan 7, 2026
One-click install
npx skills add https://github.com/simota/agent-skills --skill judge-simota
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Judge
Source: https://github.com/simota/agent-skills/tree/main/judge
Command: npx skills add https://github.com/simota/agent-skills --skill judge-simota

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates code reviews, catching bugs, security vulnerabilities, and logic errors before they reach production, ensuring higher code quality and faster development cycles.

Core Features & Use Cases

  • Automated Code Review: Performs comprehensive code reviews using codex review for PRs, pre-commits, and specific commits.
  • Bug & Security Detection: Identifies critical bugs, security flaws, and logic errors with severity classifications.
  • Use Case: Before merging a pull request, use Judge to automatically scan the code for potential issues, ensuring adherence to quality standards and security best practices.

Quick Start

Use the Judge skill to review the current uncommitted code changes.

Frequently Asked Questions about Judge

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

FAQPage Schema
How do I automate code review for pull requests to catch security vulnerabilities?

Automated code review scans pull requests using the codex review CLI to detect security vulnerabilities, bugs, and logic errors. It classifies findings by severity and provides actionable remediation suggestions before code reaches production.

Can I run automated code review on uncommitted changes in a pre-commit hook?

Yes, automated code review supports pre-commit and commit modes to analyze uncommitted changes. It checks code changes locally before they are finalized, preventing potential issues from entering the version control history.

Does automated code review work with React, Next.js, Express, and Python frameworks?

Yes, automated code review supports framework-specific reviews for React, Next.js, Express, TypeScript, Python, and Go. It analyzes code changes to detect intent misalignment and logic errors specific to these frameworks.

What is the best way to detect logic errors and bugs before merging code?

The best way to detect logic errors is using automated code review in PR mode. It identifies critical bugs and logic errors with severity classifications, ensuring adherence to quality standards and security best practices before merging.

How does the codex review CLI classify code issues and suggest fixes?

The codex review CLI classifies detected code issues by severity levels, including critical bugs and security flaws. It analyzes code changes across PR, pre-commit, and commit modes to provide actionable remediation suggestions.

Are there limitations to using automated code review for detecting intent misalignment?

Automated code review detects intent misalignment and logic errors but requires code changes to analyze. It functions across PR, pre-commit, and commit modes to provide severity classifications and actionable remediation suggestions.