codex-review

Run OpenAI Codex on code changes and compile categorized findings.

2|Updated Feb 23, 2026

Search Tags:#workflow #integration #codex #ai-assistant #code-review #pull-request #diff-review

Ready to run

npx skills add https://github.com/jonmumm/skills --skill codex-review-jonmumm

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/jonmumm/skills/tree/main/codex-review
Command: npx skills add https://github.com/jonmumm/skills --skill codex-review-jonmumm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, automated approach to performing code reviews using OpenAI Codex to catch issues and track feedback across iterations.

Core Features & Use Cases

Codex-guided review: Run Codex to review code changes and generate findings.
Iterative correction: Address findings, re-run reviews to verify fixes.
Workflow integration: Works with diff scopes (uncommitted, commit, branch diff) and can be embedded in /swarm, /nightshift contexts.

Quick Start

Provide a Codex review of the current repository changes.

Frequently Asked Questions about codex-review

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

FAQPage Schema

How do I automate code review for uncommitted changes and branch diffs?▼

You can automate a fresh-eyes code review by running OpenAI Codex on uncommitted changes, specific commits, or branch diffs. The review generates categorized, actionable findings that can be tracked and re-reviewed after iterative fixes.

Can I use OpenAI Codex to review a specific commit in my repository?▼

Yes, Codex can review a specific commit by targeting that commit's diff scope. The automated process analyzes the code changes and compiles actionable findings categorized for tracking, enabling iterative fixes and re-review to verify corrections.

What's the best way to track code review findings across multiple iterations?▼

The best way to track findings across iterations is using an automated Codex review that categorizes actionable findings. After addressing feedback, re-run the review to verify corrections and ensure iterative fixes are resolved.

Does this automated diff review work with workflow integration contexts like /swarm or /nightshift?▼

Yes, the Codex-guided code review can be embedded in /swarm and /nightshift workflow contexts. It applies to branch diffs and uncommitted changes, enabling cross-review between models and iterative fixes within automated pipelines.

How does cross-review between AI models work for pull requests?▼

Cross-review between AI models for pull requests works by running OpenAI Codex on code changes to generate fresh-eyes findings. These categorized findings enable iterative fixes, and optional re-review verifies corrections, allowing models to complement each other's review coverage.