Mails.ai — Agent Email MCP server

Read from the package source without running it. The installed server may list more.

• mails.allowlist_address

  Override suppression for one address with a >=20 char attestation.

• mails.check_suppression

  Check if an address is suppressed.

• mails.create_agentwrite action

  Create a new agent.

• mails.create_draftwrite action

  Stage a draft. Optionally pass send_at to schedule.

• mails.forward

  Forward a message to new recipients.

• mails.get_event

  Get a specific inbound event by ID.

• mails.get_received

  Get one received message including extracted reply text.

• mails.get_reputation

  Get an agent's sending-reputation health (0-1) + its 30-day bounce/complaint/reply counts. The firewall auto-suspends an agent that crosses a 0.3% complaint rate — well before the upstream provider's 0.5% line — so check this to catch an at-risk agent before it gets paused.

• mails.get_thread

  Get a full thread with all messages in chronological order.

• mails.get_usage

  Get the current billing period usage.

• mails.list_agents

  List the agents in the workspace. Returns SENDABLE agents by default (status=live); archived agents cannot send, so they are excluded unless you ask for them. Pass status="all" to see archived ones too.

• mails.list_messages

  List recent cold/first-contact inbound (message.received) for an agent — senders that are NOT replying to one of your sends.

• mails.list_received

  List recently received messages.

• mails.list_replies

  List recent reply events (reply.received) for an agent.

• mails.list_threads

  List threads (conversations) for the workspace or a specific agent.

• mails.me

  Who am I — returns the authenticated workspace, agent, tier, and API-key scopes. Call this first to confirm the connection is live.

• mails.reply

  Reply to a previously-received message. Sets In-Reply-To + Re: subject server-side.

• mails.sendwrite action

  Send an email. `agent` is OPTIONAL: omit it and the workspace's single agent is used, or one is created for you (named from `from`) — so a first send needs no setup. Pass `agent` only when the workspace has several and you must say which. The reputation firewall runs on every send: an agent auto-suspended for crossing a 0.3% complaint rate is blocked (422 agent_paused), suppressed recipients are s

• mails.send_draftwrite action

  Send a previously-created draft.

• mails.test_inbound

  SANDBOX (test key only): simulate an email arriving to one of your agents, so you can exercise the whole two-way loop — receive → prompt-injection scan → reply — with NO live mail server and no waiting. This runs the REAL reputation firewall. Read the VERDICT from top-level `quarantined` (boolean — true means we judged it an attack); the evidence sits in the `classification` block: injection_score

• Code scan5 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitystatic API keys via environment variables6/15
• Maintenanceno repository listed3/15
• Maintainer identityno repository or website to verify2/10

Mails.ai — Agent Email: common questions

Is Mails.ai — Agent Email MCP server safe?

With care: it is graded C, so read the findings first (55/100). Read the Mails.ai — Agent Email safety report

How do I install Mails.ai — Agent Email?

It runs on your machine. Copy the Claude Code, Claude Desktop or Cursor config from the install section.

Does Mails.ai — Agent Email need an API key?

Yes. The registry entry asks for MAILS_API_KEY.

Is Mails.ai — Agent Email maintained?

The latest release is v0.2.9.

What can I use instead of Mails.ai — Agent Email?

Servers from other publishers that do the same job: Atomic Mail MCP server, MCP server and Agent Email MCP server. Compare all Mails.ai — Agent Email alternatives.