---
title: "codex-in-claude"
description: "Run Codex from Claude Code for adversarial review, feedback triage, in-scope fixes, and task delegation."
canonical: https://agentpluginsdirectory.com/plugins/codex-in-claude
last-updated: 2026-09-18
---

# codex-in-claude
Run Codex from Claude Code for adversarial review, feedback triage, in-scope fixes, and task delegation.
- Slug: codex-in-claude
- Publisher: Alex Baker
- Repository: https://github.com/0xA3B/skills
- Manifest: plugins/codex-in-claude/plugin.json
- Version: 1.1.1
- License: MIT
- Category (editorial): other
- Skills: 2 (adversarial-review, using-codex-cli)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-05-02
- Repository last pushed: 2026-09-17
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/codex-in-claude
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What codex-in-claude does, in the publisher's words

Run Codex from Claude Code for adversarial code review and task delegation.

- codex-in-claude:adversarial-review: Invokes Codex as an adversarial reviewer running at the user's configured sandbox and permission defaults, has Claude triage the feedback, applies accepted in-scope fixes for current working-tree changes, validates the result, and summarizes the outcome.
- codex-in-claude:using-codex-cli: Internal contract for non-interactive Codex CLI use: command recipes, sandbox modes, sessions, structured output, and GPT-5.5 prompting guidance. The review skill and the codex agent use it; it is not meant for direct invocation.
- codex: A general-purpose proxy agent that forwards a delegated task to codex exec and returns the result. Use it to bring Codex in as a teammate for implementation, diagnosis, or research tasks, or as a subagent in multi-agent workflows. Review requests belong to the adversarial-review skill instead.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/0xA3B/skills/HEAD/plugins/codex-in-claude/README.md

## Skills

- adversarial-review: Use when the user explicitly asks Claude to use Codex to review repository code changes, the working tree, a branch, or a PR/MR, or when another loaded skill directs a Codex review pass that the user requested. Do not use for generic review requests, generic adversarial review requests, architectur…
- using-codex-cli: Internal contract for invoking the Codex CLI non-interactively. Use only when another codex-in-claude skill or the codex proxy agent needs to run codex exec, including to support a review workflow, delegate work, produce structured output, or resume or continue a prior Codex session or thread with…

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.