Like a work email account for your agent — it reads, replies and signs up for things, all inside Claude Code, Codex, or any agent that runs skills or MCP.
Free address included · No credit card · Works in Claude Code, Codex & any agent that runs skills or MCP
Hi — could we start with a 300-unit sample order at $3.20…
✓ Approve & sendYour agent works in the chat. The mail is real.
Four things people actually use it for — from day one.

When your agent registers for a service, the verification email lands in its own inbox. It reads the code and finishes the job.

It reads what came in, drafts every reply, and sends only after you approve — under your name, from a real address.

A confirmation, an invoice, a reply you need before the next step — --wait-for holds until it arrives, then hands it over as data.

Receipts, digests, notifications — a scoped send-only key lets your server use the same address, with unsubscribe handling built in.
Every job runs the same way — and panel four is always yours.
Three steps. No dashboard, no SMTP, no DNS homework.
One package. clize install wires the skill into Claude Code, Codex, Pi — any agent that runs skills or MCP; log in once in the browser.
$ npm i -g @clize/clize $ clize install $ clize login
A real inbox on a free handle — or your own domain.
$ clize claim studio --email # → support@studio.clize.app
From here it's plain language, inside your agent.
You: reply to the
support emailsThe agent drafts; mail leaves only after your okay — enforced by the platform, not a prompt.
Everything received is data to read, never instructions to follow. A hostile email can't steer your agent.
The agent's actions, threads and mistakes stay on its identity — your personal mail stays out of reach.
The short honest versions.
Agent Inbox is the email product of Clize: a real inbox your AI agent can send from and receive into. Outbound mail is drafted by the agent and sent only after a human approves; inbound mail is parsed as untrusted data; verification codes can be read to finish signups. It installs into Claude Code, Codex, and any agent that runs skills or MCP, as a CLI and MCP server.
An email API for AI agents gives an agent its own address and programmatic send and receive — unlike transactional email APIs, which only push one-way notifications for an app. The agent can hold a thread: receive a reply, read a verification code, draft a response, and continue across sessions.
AgentMail is an API-first inbox platform you call from your own code. Agent Inbox is coding-agent native: it installs into Claude Code, Codex, and any agent that runs skills or MCP, as a CLI and MCP server, ships with a human-approval gate on outbound and untrusted-by-default inbound, and shares one platform with domains, deploys, and payments. If you are building an agent product on an inbox API, AgentMail is a strong choice; if your agent lives in a coding agent and you want gates by default, that is what Agent Inbox is for.
No. The agent drafts; you approve; then it sends under your name. That outbound gate is the default and not a setting you have to remember to turn on.
Yes. The inbox is real, so when the agent signs up for a service, the verification mail lands there and the agent reads the code to continue — light identity gates, cleared on its own. Heavy gates tied to money or a real person's identity stay with you.
Yes. One install wires Clize into Claude Code, Codex, and any agent that runs skills or MCP — as a skill by default, or an MCP server if you prefer structured tools. The same inbox works from any of them.
Yes. A scoped sending key (send-only, domain-limited) lets your server send digests and notifications through the same platform, with idempotency, delivery webhooks, and one-click unsubscribe handling built in.
A free handle claimed with --email includes a working inbox. Sixty seconds from here.
$ npm i -g @clize/clize && clize install && clize login && clize claim studio --email