{"slug": "openai-ships-codex-into-claude-code-two-commands-or-four", "title": "OpenAI ships Codex into Claude Code — two commands, or four?", "summary": "OpenAI shipped an official plugin, codex-plugin-cc, that runs its Codex coding agent from inside Anthropic's Claude Code CLI. The plugin delegates to the existing Codex CLI and app server, reusing auth and configuration, and is Apache-2.0 licensed. OpenAI stated: 'It is not a separate runtime. It is Codex, just invoked from inside Claude Code.'", "body_md": "A rare thing happened on March 30, 2026: OpenAI shipped first-party tooling straight into a competitor's terminal. The result is `codex-plugin-cc`\n\n, and it changes how a two-agent workflow fits in one window.\n\n`codex-plugin-cc`\n\nis an official OpenAI plugin that runs the Codex coding agent from inside Anthropic's Claude Code CLI. It lives at [openai/codex-plugin-cc](https://github.com/openai/codex-plugin-cc), is Apache-2.0 licensed, and is effectively pure JavaScript . The notable part is the direction of travel — a frontier-model vendor packaging tooling for a rival's ecosystem.\n\nIt is not a new runtime. As OpenAI puts it in its developer-forum announcement:\n\n\"It is not a separate runtime. It is Codex, just invoked from inside Claude Code.\" — OpenAI (source:\n\n[OpenAI Developer Community])\n\nThe plugin delegates to the Codex CLI and app server already installed on your machine, reusing your existing auth and configuration . Maintenance is active: `v1.0.0`\n\nlanded March 30, 2026, and the current `v1.0.6`\n\nshipped July 8, 2026 — six patch releases in four months . This is the same Codex agent that hit more than 5 million weekly users by June 2026 , surfaced in a different terminal.\n\nThe prerequisites are modest but specific. You need Node.js 18.18.0 or later — this is a hard engine constraint declared in the plugin's `package.json`\n\n(`engines: \">=18.18.0\"`\n\n), not an advisory floor, so an older runtime will refuse to install . On the account side, either a ChatGPT subscription — the free tier qualifies — or an OpenAI API key unlocks Codex .\n\nAuth is inherited, not re-entered. If you are already logged into Codex locally, the plugin reuses that state automatically — no second login . No Codex installed yet? `/codex:setup`\n\ncan offer to run `npm install -g @openai/codex`\n\nwhen npm is present, or you can run that command manually first .\n\nThe full official path is four slash commands run inside an active Claude Code session, not the two the viral clip advertises . Work through them in order:\n\n`/plugin marketplace add openai/codex-plugin-cc`\n\n`/plugin install codex@openai-codex`\n\n`codex`\n\n, the marketplace is `openai-codex`\n\n, which is why the reference reads `codex@openai-codex`\n\n.`/reload-plugins`\n\n`/codex:*`\n\nskills never surface, and the `codex:codex-rescue`\n\nsubagent stays absent from `/agents`\n\n.`/codex:setup`\n\n`npm install -g @openai/codex`\n\n; if Codex is present but not logged in, it prompts you to run `!codex login`\n\nfrom Claude Code .Verify before you rely on it: once `/codex:setup`\n\ncompletes without complaint, `/codex:review`\n\nshould respond, and the `codex:codex-rescue`\n\nsubagent should be visible in `/agents`\n\n. If either is missing, re-run `/reload-plugins`\n\n— a skipped reload is the most common reason the command surface fails to appear.\n\nThe viral \"two commands, 30 seconds\" pitch covers only the marketplace add and the plugin install — steps 1 and 2. Skip `/reload-plugins`\n\nand the plugin is registered but dormant: its `/codex:*`\n\nslash commands never load into the running session, so nothing appears to work . Skip `/codex:setup`\n\nand the plugin never checks whether Codex is installed and authenticated — the failure is silent rather than a readable error, which is the worst kind to debug .\n\nTwo behaviors deserve caution before you lean on this in real work:\n\n`Stop`\n\nhook to fire Codex on every Claude response. OpenAI and practitioners warn this can spin up a sustained Claude/Codex loop that drains usage limits fast — treat it as optional, not a default .Keep it current, too. `v1.0.6`\n\n, released July 8, 2026, removed shell expansion for git commands — a correctness and security fix for a plugin that shells out inside your repo. Earlier builds carried Windows-compatibility and app-server readiness regressions, so pin a recent version .\n\nAfter `/codex:setup`\n\ncompletes, the plugin exposes a focused set of `/codex:*`\n\nslash commands plus a `codex:codex-rescue`\n\nsubagent visible in `/agents`\n\n. This is a command surface, not a general orchestration runtime — each command maps to one Codex action against your local checkout.\n\n| Command | What it does | Key flags |\n|---|---|---|\n`/codex:review` |\nRead-only Codex review of uncommitted changes or a branch diff |\n`--base` , `--wait` , `--background`\n|\n`/codex:adversarial-review` |\nSame target selection, but a skeptical pass that questions design decisions and assumptions, not just bugs |\n`--base` + optional focus text |\n`/codex:rescue` |\nDelegates investigation or a fix to the `codex:codex-rescue` subagent |\n`--model` (`spark` → `gpt-5.3-codex-spark` ), `--resume` , `--fresh` , `--effort`\n|\n`/codex:transfer` |\nHands the active Claude Code session into a persistent Codex thread for longer delegation (added in v1.0.5) | — |\n`/codex:status` · `/codex:result` · `/codex:cancel`\n|\nManage background jobs | — |\n\nOne practical habit: before you lean on a blocking `--wait`\n\n, run background jobs and poll them with `/codex:status`\n\nand `/codex:result`\n\n, cancelling with `/codex:cancel`\n\n— `--wait`\n\nties up your terminal until Codex returns . The concrete takeaway: install a pinned recent build, run `/codex:review`\n\nonce to confirm the subagent responds, then wire the review commands into your ship checklist — you now have Claude planning and Codex reviewing in a single terminal .\n\nNo. A ChatGPT free-tier account qualifies, and an OpenAI API key works as well . The plugin does not maintain its own credentials — it reuses whatever authentication your local Codex install already holds. If you are already logged into Codex on the machine, that auth is picked up automatically; if not, run `!codex login`\n\nfrom Claude Code once and you are set.\n\nNo. Codex usage counts against your Codex limits, and Claude usage against Claude's — there is no shared quota, no cross-account auth, and no linked billing . The plugin is a convenience bridge that invokes your locally installed Codex CLI and app server; it does not prove the two agents share context, accounts, or permissions. Watch the optional \"review gate\" in particular, since a Claude/Codex loop can drain Codex limits fast.\n\nYou most likely stopped at the two install commands and skipped `/reload-plugins`\n\n. Adding the marketplace and installing the plugin registers it, but the running session has not yet loaded the new skills. Run `/reload-plugins`\n\n, then `/codex:setup`\n\n, and the `/codex:*`\n\nslash commands plus the `codex:codex-rescue`\n\nsubagent will show up in `/agents`\n\n. This is why the \"two-command\" framing is really four.\n\n`/codex:review`\n\nruns a standard, read-only Codex pass over uncommitted changes or a branch diff against a base ref such as `main`\n\n, supporting `--base`\n\n, `--wait`\n\n, and `--background`\n\n. `/codex:adversarial-review`\n\nruns a steerable, skeptical pass that actively questions design decisions and assumptions, using the same target selection plus optional focus text. Reach for the adversarial pass when you want a pre-merge second opinion, not a routine change summary.\n\nNo. `openai/openai-developers-for-claude`\n\nis a separate plugin that bundles a Docs MCP server and API-setup skills, while `openai/codex-plugin-cc`\n\nexposes the Codex coding agent itself . They live in different repositories and serve different purposes; installing one does not give you the other. If your goal is code review and task delegation, `codex-plugin-cc`\n\nis the one you want.", "url": "https://wpnews.pro/news/openai-ships-codex-into-claude-code-two-commands-or-four", "canonical_source": "https://dev.to/creeta/openai-ships-codex-into-claude-code-two-commands-or-four-c7l", "published_at": "2026-07-22 03:53:41+00:00", "updated_at": "2026-07-22 04:04:24.582348+00:00", "lang": "en", "topics": ["artificial-intelligence", "developer-tools", "ai-products"], "entities": ["OpenAI", "Codex", "Claude Code", "Anthropic", "codex-plugin-cc"], "alternates": {"html": "https://wpnews.pro/news/openai-ships-codex-into-claude-code-two-commands-or-four", "markdown": "https://wpnews.pro/news/openai-ships-codex-into-claude-code-two-commands-or-four.md", "text": "https://wpnews.pro/news/openai-ships-codex-into-claude-code-two-commands-or-four.txt", "jsonld": "https://wpnews.pro/news/openai-ships-codex-into-claude-code-two-commands-or-four.jsonld"}}