codex

Wrap OpenAI Codex CLI with review, challenge, and consult modes.

Updated Apr 21, 2026
One-click install
npx skills add https://github.com/chriswestt/claude-skills --skill codex-chriswestt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/chriswestt/claude-skills/tree/main/gstack/codex
Command: npx skills add https://github.com/chriswestt/claude-skills --skill codex-chriswestt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenAI Codex CLI wrapper with three modes for code-related tasks: review, challenge, and consult. It enables structured workflows for code review, adversarial testing, and interactive queries with session continuity.

Core Features & Use Cases

  • Independent diff review with pass/fail gate in review mode.
  • Adversarial prompts to stress-test code in challenge mode.
  • Session-continuous questions and follow-ups in consult mode.
  • Voice triggers and seamless switching among modes for coding workflows.

Quick Start

Invoke codex with one of three modes: codex review, codex challenge, or codex consult to begin.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I run automated code review with a pass or fail gate?

Automated code review runs in review mode by analyzing independent diffs and returning a pass or fail gate based on the CLI wrapper's prompt evaluation.

Can I use adversarial prompts to stress-test my code logic?

Adversarial prompts can be applied in challenge mode to stress-test code logic and expose potential vulnerabilities within your software engineering workflow.

How do I maintain session continuity when asking follow-up coding questions?

Session continuity for follow-up coding questions is maintained in consult mode, which preserves context across interactions for continuous queries.

Do I need OpenAI Codex access to use the CLI wrapper modes?

OpenAI Codex access is required to use the CLI wrapper, along with the wrapper itself to manage mode-specific prompts for review, challenge, and consult tasks.

What is the best way to switch between code review and interactive consult workflows?

Switching between code review and interactive consult workflows is handled seamlessly by invoking codex review or codex consult commands with voice triggers.

Are there limitations when using CLI wrappers for adversarial testing of prompts?

Limitations include dependency on OpenAI Codex access and the CLI wrapper's ability to manage mode-specific prompts for adversarial testing, requiring structured inputs.