How does it handle model selection?

The skill includes logic to automatically select the standard gpt-5 model for analysis and architecture tasks, while switching to gpt-5-codex for file editing and refactoring.

Do I need the Codex CLI installed to use this skill?

Yes, this skill requires the Codex CLI (v0.48.0 or higher) to be installed and authenticated on your system for the execution helpers to function.

What does the Codex CLI Bridge do?

It creates a functional connection between Claude Code and OpenAI's Codex CLI, enabling synchronized documentation and the ability to run Codex-specific commands directly through Claude Code.

Does it duplicate my project files when generating documentation?

No, it uses a reference-based approach for AGENTS.md generation, meaning it only creates links to existing files rather than duplicating content.

Is it safe to run code editing tasks via this bridge?

Yes, the skill enforces specific sandbox modes, such as 'workspace-write' for edits and 'read-only' for analysis, to ensure security and prevent unauthorized system access.

Codex CLI Bridge

Name: Codex CLI Bridge
Author: chaorenex1

bychaorenex1

生产力与工作流

Bridges Claude Code and OpenAI Codex CLI to enable seamless task execution and documentation synchronization between both toolsets.

关于

The Codex CLI Bridge skill establishes a robust link between Claude Code and OpenAI’s Codex CLI, allowing developers to leverage the specialized strengths of both environments. It automates the generation of project-specific AGENTS.md documentation from existing CLAUDE.md files using a reference-based approach that avoids file duplication. Beyond documentation, it provides Python-based execution helpers for complex tasks like security analysis and code refactoring, ensuring safety through automated environment validation and strict sandbox controls.

主要功能

Safe Python execution wrappers for specialized codex exec commands
Automated CLAUDE.md to AGENTS.md translation via reference-based links
Intelligent model selection between gpt-5 and gpt-5-codex based on task
0 GitHub stars
Automated project initialization and Codex environment validation checks
Comprehensive mapping of Claude Code Skills for Codex CLI users

使用场景

Executing complex code refactoring tasks using Codex CLI specialized models
Synchronizing documentation between Claude Code and Codex-based agent environments
Automating security vulnerability analysis with read-only sandbox execution

关于

主要功能

Safe Python execution wrappers for specialized codex exec commands
Automated CLAUDE.md to AGENTS.md translation via reference-based links
Intelligent model selection between gpt-5 and gpt-5-codex based on task
0 GitHub stars
Automated project initialization and Codex environment validation checks
Comprehensive mapping of Claude Code Skills for Codex CLI users

使用场景

Executing complex code refactoring tasks using Codex CLI specialized models
Synchronizing documentation between Claude Code and Codex-based agent environments
Automating security vulnerability analysis with read-only sandbox execution