Communication
Overview
Try it free for 7 days: a sandbox mailbox, no card. New here? Start at aimailmcp.com/docs, where the Start here box has an interactive demo and a getting-started guide. AI Mail MCP gives an AI agent a real email inbox over MCP. Connect it to Claude and your agent can read one unified inbox across its mailboxes, search mail, wait for a sign-up code or magic link and pull it out, sort recent mail into buckets (needs a reply, bills and receipts, newsletters, notifications), and send, reply in-thread and forward. It can also add a domain and get the exact DNS records to set, create mailboxes and aliases, and set up plain-language labels and rules. Mail from outside senders is untrusted, and the server treats it that way. Every inbound message is scored for prompt injection, and a message scored as one has its body withheld until the agent explicitly asks for it. Content is returned inside an envelope that marks it as data. Outgoing mail passes a guard that checks for credentials, first-time recipients, look-alike domains and misleading links, and holds a credential going to a new recipient until an account owner approves it. AI Mail's own sign-in codes are never readable by an agent. You stay in control. Sign-in is OAuth 2.1 with an emailed code. A connection can only read mail unless you tick "allow sending" or "allow managing the account" on the consent page. Every tool that changes something accepts dry_run, every tool call is written to an audit log, and sending is capped per day. After the trial, pricing is prepaid and per use, with no subscription; a person pays a Stripe Checkout link and the agent never handles card details. The 7-day sandbox gives one mailbox that receives mail and sends only to your own verified email and other sandbox mailboxes.
