About MachineMail

MachineMail is an email API for AI agents. It gives software that reads and writes its own mail a real inbox: create one with one API call, send and receive from it, and get every inbound message back as structured JSON with an agent-safe body and a verdict on whether it can be trusted.

What MachineMail does

• Free addresses. Inboxes on the shared machinemail.dev domain cost nothing, on every plan. Make one per agent, per task or per sign-up flow. Ephemeral inboxes expire on their own, and a catch-all domain turns every address into an inbox the moment mail arrives.
• A trust layer on every message. SPF, DKIM and DMARC, sender history, hidden text and zero-width characters, link and attachment checks, and a classifier that flags prompt injection, phishing and requests for credentials. Your agent reads the cleaned body and the verdict, not raw HTML from a stranger.
• AgentMail-compatible. The /v0 API follows AgentMail's paths, field names and enums, so an existing client switches by changing the base URL and the API key.
• Built for agents. A hosted MCP server, real-time events over webhooks, Server-Sent Events and WebSockets, and auth.md self-registration: an agent can sign itself up and get a sandbox inbox and an access token, and a human claims it later.

Why a trust layer

An inbox is an open door. Anyone who knows the address can put text in front of your agent, and the agent reads that text with the same attention it gives your instructions. MachineMail checks every inbound message before it is stored and attaches a trust report that says who really sent it and whether anything in it is trying to manipulate the reader. Your agent acts on the trust level, never on what the sender says. The trust layer docs explain every check.

How to start

• Free plan. Create an account and start on Free, at $0. See Pricing for what each plan includes.
• Self-serve API keys. Make and revoke keys yourself in the dashboard. The quickstart goes from sign-up to a sent message.
• A sandbox for agents. An agent that follows auth.md gets a sandbox inbox and a working credential without a human creating a key first, then shows you a 6-digit code to claim it. Agent self-registration describes the sandbox limits.
• Every way in. The REST API reference, the OpenAPI document, the MCP server at https://mcp.machinemail.dev/mcp, and the CLI.

Find us

• Docs: machinemail.dev/docs
• GitHub: github.com/machinemail, with the agent skills and the agent plugin
• X: @machinemail_dev
• Service status: status.machinemail.dev
• Questions: see Contact