Xiaolai Codex Toolkit For Claude
xiaolai-codex-toolkit-for-claude is a Claude Code plugin for the Ship phase that connects OpenAI Codex via MCP to audit, implement, verify, and debug your codebase.
by xiaolai · github.com/xiaolai/codex-toolkit-for-claude
Bridge Claude Code to OpenAI Codex over MCP so you can audit, implement, verify, and debug code with a second model without leaving the agent.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install xiaolai-codex-toolkit-for-claude@xiaolai/codex-toolkit-for-claudeBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:xiaolai/codex-toolkit-for-claude") and the paste-ready config with skillselion.get_install_config("plugin:xiaolai/codex-toolkit-for-claude").
What it does
xiaolai-codex-toolkit-for-claude is a one-plugin marketplace that wires OpenAI Codex into Claude Code through the Model Context Protocol. Solo builders install it when Claude alone is not enough for disciplined review, when they want a second model to implement or verify changes, or when debugging needs a dedicated Codex-backed pass without switching tools. The described surface area—audit, implement, verify, and debug—covers late build integration work and ship-time review, with debug extending into operate. You should expect MCP configuration, valid Codex access, and Claude Code as the host. The bundle is intentionally narrow: it is an integration toolkit, not a full agent skills library. Use it to add measurable review and verification steps to your existing Claude Code hooks and commands rather than as a standalone product framework.
Highlights
- Single-plugin marketplace focused on OpenAI Codex MCP for Claude Code.
- Supports audit, implement, verify, and debug flows in one toolkit.
- Positions as code-review and verification layer, not a replacement IDE.
- Developer-tools category listing with explicit MCP and Codex keywords.
- Designed for dual-model workflows: Claude orchestrates, Codex executes reviews or patches.
Why builders use it
Staying inside Claude Code while still getting Codex-grade review, implementation checks, and debug help usually means fragile copy-paste or context switching.
After registration, Claude Code can call Codex through MCP for structured audit, verify, implement, and debug steps on the same repository context.
At a glance
- Type - Plugin in LLM Integration.
- Adoption - 0 installs, 21 stars, 0 votes.
FAQ
Who is xiaolai-codex-toolkit-for-claude for?
Developers who run Claude Code as their primary agent and want OpenAI Codex available for audit, implementation help, verification, and debugging through MCP.
When should I use xiaolai-codex-toolkit-for-claude?
Use it before release when you need an extra model pass on reviews, when implementing risky changes, or when debugging with Codex-backed analysis in the same workflow.
How do I add xiaolai-codex-toolkit-for-claude to my agent?
Install the xiaolai/codex-toolkit-for-claude plugin from its marketplace entry, configure the Codex MCP server per the repo README, and enable the plugin in Claude Code settings.
Comments
Share how you use xiaolai-codex-toolkit-for-claude, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.