About AgenticEmail
AgenticEmail is a hosted email service for AI agents that need to work with real inboxes. Agents can provision inboxes and send, read, and reply to messages through native MCP tool calls.
Key features
- 01Provision and list inboxes
- 02Browse threads and message history
- 03Read parsed messages
- 04Send email from an inbox
- 05Reply within an existing thread
- 06Scope access with API keys
Use cases
- 01Create an inbox for an onboarding flow
- 02Send welcome emails to new contacts
- 03Monitor incoming replies to an email thread
MCP support
AgenticEmail provides a hosted remote MCP server over Streamable HTTP, with tools for inbox creation, message reading, sending, and replies. Scoped API keys limit which inboxes an agent can access.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://api.agenticemail.dev/mcpclaude mcp add --transport http agenticemail https://api.agenticemail.dev/mcp{"mcpServers":{"agenticemail":{"type":"http","url":"https://api.agenticemail.dev/mcp"}}}Open in Cursor{"servers":{"agenticemail":{"type":"http","url":"https://api.agenticemail.dev/mcp"}}}Open in VS Code{"mcpServers":{"agenticemail":{"type":"http","url":"https://api.agenticemail.dev/mcp"}}}{"mcpServers":{"agenticemail":{"serverUrl":"https://api.agenticemail.dev/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://api.agenticemail.dev/mcpAuth requiredTools
9| Parameter | Type | Required | Description |
|---|---|---|---|
| display_name | string | No | |
| domain | string | No | |
| username | string | No |
Details
From the project README
AgenticEmail MCP Server
Hosted email for AI agents over the Model Context Protocol
Website · Docs · MCP Registry: dev.agenticemail/mcp · Discord
AgenticEmail ships a hosted MCP server so any MCP-compatible agent (Claude, Cursor, Codex, custom agents) can create inboxes and send, receive, and reply to real email as native tool calls - no custom glue code.
The server is remote (streamable HTTP). There is nothing to install or run locally.
Connect
{
"mcpServers": {
"agenticemail": {
"url": "https://api.agenticemail.dev/mcp",
"headers": {
"Authorization": "Bearer ${AGENTICEMAIL_API_KEY}"
}
}
}
}
Get an API key at agenticemail.dev - free while you build. Scoped keys work here too, so an agent only sees the inboxes you grant it.
Tools
| Tool | Description |
|---|---|
create_inbox | Provision a new addressable inbox |
list_inboxes | List inboxes the key can access |
list_threads / get_thread | Browse conversations and load full thread history |
list_messages / get_message | Read parsed inbound and outbound messages |
send_message | Send a message from an inbox |
reply_to_message | Reply within an existing thread |
Example
Create an inbox for the onboarding flow, send a welcome email to [email protected], then watch for their reply.
The agent calls create_inbox, then send_message, and reacts to the reply through your webhook or WebSocket stream.
Beyond MCP
The same platform is available as a REST API, with webhooks, real-time events, custom domains, SMTP compatibility, and official TypeScript and Python SDKs.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.