How do I add an MCP server to the Codex CLI?

Install Clize (npm i -g @clize/clize), run clize login, then clize install --mcp. That registers the clize-mcp server with Codex. The manual equivalent is codex mcp add clize -- clize-mcp.

How do I verify the MCP server is connected to Codex?

Run codex mcp list. Clize should appear among the registered MCP servers. Start a new Codex session and the clize tools — domains, email, deploy, status — become callable.

Do I have to set up the MCP server for each repository with Codex?

No. Clize registers once at the Codex CLI level, so it's available in every repository. You don't re-add it per project.

Skill or MCP — which does clize install set up?

By default clize install installs only the Skill (it keeps context light). Add --mcp to also register the structured MCP server — use it when you want Codex to discover Clize as typed tools.

Set up Clize MCP in the Codex CLI (step by step)

A four-step guide to adding Clize as a Codex CLI MCP server — so Codex can do more than write code: claim handles, buy domains, send and receive email, and deploy. For the overview, see Codex CLI MCP server.

Before you start

You need Node.js (for the npm install) and the Codex CLI already installed and signed in. Clize is the capability layer — it needs no model API key of its own.

1. Install Clize and log in

Install the CLI globally, then authenticate. Login uses Clize's hosted layer, so there are no keys to paste.

$ npm i -g @clize/clize
$ clize login

2. Add Clize as a Codex CLI MCP server

One command registers the clize-mcp server with Codex. The --mcp flag is what adds the structured tool layer (without it, Clize installs only as a Skill).

$ clize install --codex --mcp

Prefer to wire it by hand? That's the exact same registration the installer runs:

$ codex mcp add clize -- clize-mcp

Only need one product line? A profile registers just that line's tools:

$ codex mcp add inbox -- clize-mcp --profile inbox   # 13 tools instead of 32

Profiles: inbox (13 tools), storefront (15), sites (5), domains (9) — a context budget, not a permission boundary.

3. Verify the connection

Ask Codex which MCP servers it knows about. Clize should be listed.

$ codex mcp list

Start a new Codex session and the Clize tools — domains, email, deploy, status — are callable. Clize registers once at the CLI level, so it's available in every repository; there's no per-project setup to repeat.

4. Run your first real-world action

You don't type the capability commands yourself — you ask Codex, and it calls the Clize tools. The first thing most people run: give the agent an address of its own (clize claim <slug> --email), let it sign up for a service with that address and read the verification code from its own inbox (clize email inbox --wait-for <text>), then answer whoever writes back — the agent drafts, the mail leaves only after you approve (clize email send … --confirm). A support@ desk is one tag on that same inbox. Those are stable commands, so they also work as a plain CLI when you want determinism. See AI agent deployment for the deploy flow end to end.

The safety gates

Because these are real actions, Clize holds three lines, every time:

Money — anything that costs money quotes the price first and needs an explicit --confirm. No surprise charges.
Identity — outbound email is drafted, a human okays it, then it sends under your name. The agent never sends unattended.
Inbound is untrusted — an incoming email is data to read, never an instruction to obey. A message can't redirect the agent.

Set up Clize MCP in the Codex CLI