mail-mcp
A self-hosted, multi-mailbox MCP server that exposes IMAP/SMTP operations as tools for AI agents, enabling email management like listing folders, reading, searching, sending, and moving messages via natural language.
README
mail-mcp
Self-hosted, provider-agnostic mail service. Point it at any mailbox's IMAP/SMTP credentials and it exposes:
- REST API mirroring the common mailbox operations: list folders, list/get/search messages, send, flag, move, delete, quota (best-effort).
- Webhooks (
message.received) via a persistent IMAP IDLE watcher — fires an HTTP POST (with a per-webhook bearer secret) whenever new mail arrives, retried with exponential backoff (up to 6 attempts) if the receiving endpoint fails. - MCP server (Streamable HTTP,
POST /mcp) exposing the same operations as tools, so any MCP-compatible agent can use it directly.
Multi-mailbox from the start: add as many accounts as you want to config/accounts.json.
Request bodies on the REST API are validated with Zod (lib/schemas.js) — invalid
payloads get a 422 ERR_VALIDATION_FAILED with field-level errors instead of reaching
the mail layer. Webhook subscriptions persist in SQLite (config/webhooks.sqlite,
via better-sqlite3) instead of a flat JSON file. Messages larger than
MAIL_MCP_MAX_MESSAGE_SIZE (default 25MB) are returned with bodyTruncated: true
and no parsed body/attachments, to avoid loading huge messages into memory.
Setup
npm install
cp .env.example .env # fill in MAIL_MCP_API_TOKEN and one MAILMCP_PASS_<ID> per account
Copy config/accounts.json.example to config/accounts.json (gitignored — no
passwords in this file either, those live only in .env):
[
{
"id": "example",
"user": "user@example.com",
"imap": { "host": "imap.example.com", "port": 993, "secure": true },
"smtp": { "host": "smtp.example.com", "port": 465, "secure": true }
}
]
The password for account id: "example" is read from MAILMCP_PASS_EXAMPLE.
node index.js
Runs on 127.0.0.1:4900 by default (see MAIL_MCP_PORT). Put it behind a
reverse proxy with TLS if you need to reach it from outside the host.
systemd
An example unit is in deploy/mail-mcp.service — copy it to
/etc/systemd/system/, adjust paths, and it loads secrets from /root/mail-mcp/.env
via EnvironmentFile.
Auth
Every REST and MCP request requires Authorization: Bearer <MAIL_MCP_API_TOKEN>.
Tests
npm test # run once
npm run test:coverage # run with coverage report
Coverage covers the pure/logic modules (lib/schemas.js, lib/accounts.js,
lib/webhooks-store.js). lib/mailclient.js and lib/idle-watcher.js talk to
real IMAP/SMTP servers and are exercised through manual end-to-end testing
instead of unit tests — they're excluded from the coverage badge so it isn't
misleading.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.