codex

Coordinate Claude and OpenAI Codex for collaborative code creation, review, and debugging.

13|4|Updated Feb 15, 2026
One-click install
npx skills add https://github.com/ggombee/code-forge --skill codex-ggombee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/ggombee/code-forge/tree/main/skills/codex
Command: npx skills add https://github.com/ggombee/code-forge --skill codex-ggombee

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai_codex, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill eliminates the need for repetitive code review and manual debugging, allowing you to focus on your high-level design and strategic tasks.

Core Features & Use Cases

  • Code Implementation & Review: Claude drafts code, while Codex reviews for errors and inefficiencies.
  • Collaborative Modes: Choose between Solo+Review, Sequential, or Parallel to best fit your workflow.
  • Role Segregation: Utilizes Claude for design and high-level decision making, while Codex handles precise implementation and testing.
  • End-to-End Integration: Seamlessly integrates with MCP (Multifunctional Coding Platform) server.

Quick Start

Run the 'codex' command with a specific task to begin collaborative coding.

Frequently Asked Questions about codex

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

FAQPage Schema
How does collaborative coding between Claude and OpenAI Codex work?

Collaborative coding pairs Claude with OpenAI Codex to draft, review, and debug code blocks by segregating roles: Claude handles high-level design and decisions, while Codex executes precise implementation and testing.

What modes can I use for code review and debugging with Claude and Codex?

Code review and debugging modes include Solo+Review, Sequential, and Parallel, allowing you to tailor the collaborative workflow between Claude and Codex to best fit your specific coding task requirements.

Do I need a Python environment to use OpenAI Codex for code debugging?

Yes, you need a Python environment configured for CLI headless mode, as the integration requires Python scripting to execute OpenAI Codex commands for code debugging and implementation.

Can I use this collaborative coding skill with a Multipurpose Coding Platform server?

Yes, the collaborative coding skill is optimized for end-to-end integration with a Multipurpose Coding Platform (MCP) server, allowing seamless code creation, review, and debugging operations.

What is the best way to automate repetitive code review and manual debugging?

Automating repetitive code review and manual debugging is best achieved by running the 'codex' command with a specific task, letting Claude draft the code while Codex automatically reviews it for errors and inefficiencies.