skills.rest
tommylower
tommylowerCOMMUNITY
Explore all skills by this author →
Website𝕏@tommylowerGitHub

codex-review

Configures the Codex plugin for cross-model code review inside Claude Code.

2|Updated Apr 15, 2026
Search Tags:#code review#claude code#codex#plugin#task delegation#cross-model#adversarial review
One-click install
npx skills add https://github.com/tommylower/cortex --skill codex-review-tommylower
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-review
Source: https://github.com/tommylower/cortex/tree/main/agent-workflows/codex-review
Command: npx skills add https://github.com/tommylower/cortex --skill codex-review-tommylower

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @openai/codex.

What problem does it solve? Single-model code review can miss design flaws and risky decisions. This Skill sets up OpenAI's Codex plugin inside Claude Code so a second model reviews your changes, challenges design decisions, and takes over tasks when Claude gets stuck. ## Core Features & Use Cases - Cross-Model Code Review: Run /codex:review on uncommitted changes or branch diffs for a read-only second opinion from Codex. - Adversarial Review: Use /codex:adversarial-review with free-form focus text to pressure-test auth changes, infra scripts, migrations, and complex state management before shipping. - Task Delegation: Hand off stuck tasks with /codex:rescue, optionally resuming previous runs or switching to a faster model. - Background Job Management: Track, retrieve, and cancel long-running reviews with /codex:status, /codex:result, and /codex:cancel. - Use Case: Before merging a branch that changes your caching layer, run /codex:adversarial-review --base main challenge whether this caching strategy handles invalidation correctly to get a focused critique from a different model. ## Quick Start Ask the agent to install the Codex plugin from the openai/codex-plugin-cc marketplace, run /codex:setup to verify login, then run /codex:review on your uncommitted changes.

Frequently Asked Questions about codex-review

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

FAQPage Schema
How do I set up Codex code review inside Claude Code?▼

Add the marketplace with /plugin marketplace add openai/codex-plugin-cc, install with /plugin install codex@openai-codex, then run /codex:setup to verify the Codex CLI is installed and logged in. Test with /codex:review --background.

How do I get a second AI model to review my code changes?▼

Run /codex:review to check uncommitted changes or /codex:review --base main to review a branch diff. The review is read-only and runs the Codex model against your changes without leaving your Claude Code session.

What is the difference between codex review and adversarial review?▼

Standard /codex:review checks code quality on diffs, while /codex:adversarial-review challenges design decisions and accepts free-form focus text. Use adversarial review for auth changes, infra scripts, migrations, and complex state management.

Does the Codex plugin require a paid ChatGPT subscription?▼

No, any ChatGPT subscription including Free works, or you can use an OpenAI API key. You also need Node.js 18.18+ and the Codex CLI installed globally via npm install -g @openai/codex.

Should I enable the Codex review gate for automatic reviews?▼

Only when actively monitoring. The review gate auto-reviews every time Claude finishes and blocks completion on issues, but it creates a Claude/Codex loop that can drain usage limits quickly.

What can I do when Claude gets stuck on a coding task?▼

Use /codex:rescue followed by a task description to hand the problem to Codex for a different model's approach. Add --resume to continue the last run or --model spark for a faster, cheaper model.

Related Skills

Explore More Software Engineering Skills→
PsychQuantPsychQuant

codex-docs-guide

Query OpenAI Codex CLI source code and documentation for configuration and usage details.

Official
Intermediate
kairyoukairyou
180

at-review

Review code diffs and pull requests for bugs, regressions, and convention violations.

Community
Advanced
kenkenmainkenkenmain
1

swarm

Coordinate a six-phase A0-A5 swarm pipeline with agent teams and dual-channel communication.

Official
Advanced
eprouvezeeprouveze

second-opinion

Automate cross-LLM code reviews and adversarial testing via Codex CLI.

Community
Advanced