codex

Generate a reference-based AGENTS.md from CLAUDE.md for Claude Code projects.

3|Updated Oct 15, 2025
One-click install
npx skills add https://github.com/chaorenex1/coding-agent-workflow --skill codex-chaorenex1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex
Source: https://github.com/chaorenex1/coding-agent-workflow/tree/main/当前/skills/codex
Command: npx skills add https://github.com/chaorenex1/coding-agent-workflow --skill codex-chaorenex1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the tedious and time-consuming processes of complex code analysis, large-scale refactoring, and automated code generation. It allows you to delegate these tasks to an AI agent, providing structured output and the ability to resume sessions for iterative development.

Core Features & Use Cases

  • Complex Code Analysis: Gain deep insights into codebases, identify issues, and understand logic across multiple files using AI.
  • Automated Refactoring & Generation: Delegate large-scale code changes, refactoring, and new code generation with built-in safety controls.
  • Session Management: Resume previous AI conversations to maintain context and iterate on tasks efficiently, ensuring continuity in complex projects.
  • Use Case: You need to refactor a legacy module for performance. Use Codex to analyze the module, propose changes, and then iteratively apply those changes, resuming the session as needed to refine the output.

Quick Start

Use the codex skill to explain the file 'src/main.ts'.

Frequently Asked Questions about codex

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

FAQPage Schema
How do I automate code analysis and refactoring with AI?

Code analysis and refactoring can be automated by delegating tasks to an AI agent that examines multiple files, identifies issues, and proposes structured changes. This Skill generates reference-based documentation and applies transformations while maintaining session context for iterative refinement across your codebase.

Can I resume AI sessions to continue complex code tasks?

Session management enables you to resume previous AI conversations, preserving context and maintaining continuity. This allows iterative development on large refactoring projects or code generation tasks without losing the agent's understanding of your codebase structure and requirements.

What does it mean to generate AGENTS.md from CLAUDE.md?

Generating AGENTS.md from CLAUDE.md eliminates documentation duplication by creating a reference-based agent guide derived from your primary Claude documentation. This produces CLI usage examples, workflow translations, and MCP integration metadata automatically from your existing project structure.

Does this work with existing Claude Code projects?

Yes. This Skill applies to Claude Code projects containing CLAUDE.md, .claude/skills, .claude/agents, and related documentation folders. It validates your environment, auto-initializes missing CLAUDE.md, analyzes your project structure, and generates output using template-driven processes.

How do I delegate large-scale code generation safely?

Large-scale code generation includes built-in safety controls through structured output validation and the ability to review and iterate on changes before applying them. Session resumption lets you refine proposals incrementally, ensuring generated code meets your requirements before integration.

What are the prerequisites for using this Skill?

Your project must contain CLAUDE.md and .claude/ directories with skills and agents subdirectories. The Skill handles environment validation and auto-initialization; if CLAUDE.md is missing, it generates one automatically before proceeding with analysis and documentation generation.