記事一覧へ
Claude Code向けCodexプラグインのご紹介
すでにClaude Codeを使っている方なら、このCodexプラグインを使えば同じワークフローにCodexを簡単に組み込むことができます。
主な用途は3つです: 通常のCodexレビュー、より批判的な敵対的レビュー、そして別のエージェントによる二次チェックを望む際のCodexへの作業引き渡し。
セットアップ:
/plugin marketplace add openai/codex-plugin-cc
/plugin install codex@openai-codex
/codex:setup
GitHub: https://github.com/openai/codex-plugin-cc
必要要件: ChatGPTサブスクリプション（無料プランを含む）またはOpenAI APIキー、およびNode.js 18.18以降。
主なコマンド:
/codex:review — 標準的な読み取り専用のCodexレビューを実行
/codex:adversarial-review — 挑戦的なレビューを実行。実装を疑問視させたい場合に使用
/codex:rescue — タスクをCodexに直接引き渡す
長時間のジョブはバックグラウンドで実行でき、/codex:status、/codex:result、/codex:cancelで管理できます。
最大の価値は、Claude Codeを離れずに別のエージェントから本格的な二次レビューを得られることです。基本パターンとして、/codex:reviewをデフォルトの二次チェックとして使い、重要な変更には/codex:adversarial-reviewを、スレッドが行き詰まった場合は/codex:rescueを使うことを推奨します。
claude-setupclaude-workflowai-thinking
Claude Code向けCodexプラグイン登場
♥ 1,528↻ 202
原文を表示 / Show original
If you already use Claude Code, this Codex plugin gives you a simple way to pull Codex into the same workflow.
It is useful for three things: a normal Codex review, a more skeptical adversarial review, and handing work off to Codex when you want a second pass from a different agent.
Setup
bash
/plugin marketplace add openai/codex-plugin-cc
/plugin install codex@openai-codex
/codex:setup
Check out the plugin here: https://github.com/openai/codex-plugin-cc
You need a ChatGPT subscription, including Free, or an OpenAI API key, plus Node.js 18.18 or later.
If Codex is not installed yet:
bash
npm install -g @openai/codex
If Codex is installed but not authenticated:
bash
!codex login
Core commands
/codex:review runs the standard read-only Codex review.
/codex:adversarial-review runs a steerable challenge review. Use it when you want Codex to question the implementation, not just inspect it.
/codex:rescue hands a task to Codex directly.
Longer jobs can run in the background, with /codex:status, /codex:result, and /codex:cancel to manage them.
A simple first run:
When to use it
bash
/codex:review --background
/codex:status
/codex:result
The main value here is getting a real second pass from a different agent without leaving Claude Code.
A good default pattern is simple:
Run /codex:review for all things
Run /codex:adversarial-review on anything high stakes
Use /codex:rescue when a thread stalls or you want Codex to take over
Adversarial review is especially useful for migrations, auth changes, infra scripts, refactors, and anything where the danger is hidden assumptions rather than obvious syntax mistakes.
How it works
The plugin delegates through the local Codex CLI and Codex app server. So it uses the same local auth, config, environment, and MCP setup you already have with Codex.
That is why it feels lightweight. It is not a separate runtime. It is Codex, just invoked from inside Claude Code.
Optional review gate
You can also enable a review gate:
bash
/codex:setup --enable-review-gate
That can stop Claude Code from exiting before a Codex review has run.
Useful, but something to use carefully. It can create a long Claude/Codex loop and burn through usage limits quickly.
Bottom line
This plugin is a simple way to keep your Claude Code workflow and still use Codex where Codex is strong.
Install it, run /codex:setup, use /codex:review as your default second pass, and reach for /codex:adversarial-review or /codex:rescue when you want more than a routine review.
Check out the plugin here: https://github.com/openai/codex-plugin-cc