codex

Coordinate Claude and Codex pair programming sessions via MCP server.

1|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/moveonha/pixel-ai-studio --skill codex-moveonha
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/moveonha/pixel-ai-studio/tree/main/.claude/skills/codex
Command: npx skills add https://github.com/moveonha/pixel-ai-studio --skill codex-moveonha

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables Claude and Codex to perform paired programming effectively, coordinating strengths to split complex tasks into collaborative workflows and manage sessions across turns.

Core Features & Use Cases

  • Coordinated Claude + Codex pair programming with an MCP server for task delegation and synchronization.
  • Multi-turn sessions with thread_id persistence across working directories and model contexts.
  • Supports collaboration patterns: Solo+Review, Sequential, and Parallel for different project needs.
  • Includes structured guidance, templates, and safety checks to ensure quality reviews and safe outputs.

Quick Start

Log in to Codex CLI with codex login and initiate your first session by calling the codex pair programming tool with an initial prompt.

Frequently Asked Questions about codex

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

FAQPage Schema
How does AI-assisted pair programming coordinate two models for coding tasks?

AI-assisted pair programming coordinates two models by splitting complex tasks into collaborative workflows, delegating design and precise coding across multi-turn sessions to return completed, reviewed results.

What is the best way to manage multi-turn coding sessions across different directories?

Managing multi-turn coding sessions across directories requires an MCP server and CLI tool that utilize thread_id persistence, allowing you to maintain model context and synchronize task delegation across turns.

Do I need a Codex MCP server to run collaborative code reviews?

Yes, you need the Codex MCP server and codex CLI to manage multi-turn sessions, provide structured guidance, and perform coordinated code reviews within collaborative workflows.

What collaboration patterns can I use for AI pair programming?

You can use Solo+Review, Sequential, and Parallel collaboration patterns to delegate tasks, synchronize model contexts, and ensure quality reviews for different project needs.

How do I start a pair programming session with Claude and Codex?

To start a pair programming session, log in to the Codex CLI with codex login, then initiate your first session by calling the pair programming tool with an initial prompt.

When should I use sequential versus parallel collaboration patterns for code implementation?

Use sequential patterns for step-by-step implementation dependencies and parallel patterns for independent task execution, ensuring safety checks and structured templates guide the coordinated creative coding.