myclaude/skills/codex/SKILLS.md

name, description

name	description
codex	Execute Codex CLI for code analysis, refactoring, and automated code changes. Use when you need to delegate complex code tasks to Codex AI with file references (@syntax) and structured output.

Codex CLI Integration

Overview

Execute Codex CLI commands and parse structured JSON responses. Supports file references via @ syntax, multiple models, and sandbox controls.

When to Use

• Complex code analysis requiring deep understanding
• Large-scale refactoring across multiple files
• Automated code generation with safety controls
• Tasks requiring specialized reasoning models (o3, gpt-5)

Usage

bash scripts/codex.sh "<task>" [model] [working_dir]

Timeout: Set timeout: 7200000 (2 hours) in Bash tool for long tasks.

Parameters

• task (required): Task description, supports @file references
• model (optional): Model to use (default: gpt-5-codex)
  • gpt-5-codex: Default, optimized for code
  • gpt-5: Fast general purpose
  • o3: Deep reasoning
  • o4-mini: Quick tasks
  • codex-1: Software engineering
• working_dir (optional): Working directory (default: current)

Return Format

Extracts agent_message from Codex JSON stream:

Agent response text here...

Error format:

ERROR: Error message

Examples

Basic code analysis:

bash scripts/codex.sh "explain @src/main.ts"

Refactoring with specific model:

bash scripts/codex.sh "refactor @src/utils for performance" "gpt-5"

Multi-file analysis:

bash scripts/codex.sh "analyze @. and find security issues" "gpt-5-codex" "/path/to/project"

Quick task:

bash scripts/codex.sh "add comments to @utils.js" "gpt-5-codex"

Notes

• Runs with --dangerously-bypass-approvals-and-sandbox for automation
• Uses --skip-git-repo-check to work in any directory
• Streams progress, returns only final agent message
• Requires Codex CLI installed and authenticated
• Use timeout: 7200000 (2 hours) for complex tasks that may take longer