About gmail
gmail is an MCP server that lets AI agents read, search, and send email through Gmail. It supports inbox listing and contextual replies based on message history.
Key features
- 01Search with Gmail query syntax
- 02Read full email messages
- 03List recent inbox messages
- 04Send email with CC and BCC
Use cases
- 01Triage new messages and identify urgent email
- 02Search past conversations for context before replying
- 03Draft and send routine communications on a user's behalf
MCP support
Exposes Gmail search, reading, inbox listing, and sending tools to MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @striderlabs/mcp-gmailGOOGLE_CLIENT_IDGoogle OAuth 2.0 Client ID from Google Cloud ConsoleGOOGLE_CLIENT_SECRETSecretGoogle OAuth 2.0 Client SecretGOOGLE_REFRESH_TOKENSecretGoogle OAuth 2.0 Refresh Token for the user's Gmail accountLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
4Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| cc | string | No | CC recipients (comma-separated, optional) |
| to | string | Yes | Recipient email address |
| bcc | string | No | BCC recipients (comma-separated, optional) |
| body | string | Yes | Email body content (plain text) |
| subject | string | Yes | Email subject line |
Details
From the project README@striderlabs/mcp-gmail
MCP server that gives AI agents the ability to send, read, and search email via Gmail.
For Agents
Agents use this connector to handle email autonomously on behalf of their human — reading important messages, searching for information, composing thoughtful replies, and sending communications.
Example agent workflows:
- Triage emails: Agent reads new messages, summarizes important ones, and flags urgent items for user attention
- Answer routine questions: Agent searches email history to find information, composes replies with context from past conversations
- Send on behalf: Agent drafts and sends professional emails based on user intent ("Let the team know I'll be late"), then logs a copy to the conversation
Usage example:
User: "Has anyone replied to my email about the Q2 budget proposal?"
Agent:
gmail_search→"to:me subject:budget"gmail_read→ Read each message- Return summary: "Sarah replied yes on Monday. Mark asked for revisions. No word from Finance yet."
Then agent might automatically draft a follow-up to Finance and ask for approval before sending.
Installation
npx @striderlabs/mcp-gmail
Configuration
Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"strider-gmail": {
"command": "npx",
"args": ["-y", "@striderlabs/mcp-gmail"],
"env": {
"GOOGLE_CLIENT_ID": "your-client-id",
"GOOGLE_CLIENT_SECRET": "your-client-secret",
"GOOGLE_REFRESH_TOKEN": "your-refresh-token"
}
}
}
}
Available Tools
gmail_send
Send an email from the user's Gmail account.
Parameters:
to(required): Recipient email addresssubject(required): Email subject linebody(required): Email body contentcc(optional): CC recipientsbcc(optional): BCC recipients
gmail_search
Search emails using Gmail search syntax.
Parameters:
query(required): Gmail search query (e.g., "from:[email protected]")maxResults(optional): Maximum results (default: 10)
gmail_read
Read the full content of a specific email.
Parameters:
messageId(required): The Gmail message ID
gmail_list
List recent emails from the inbox.
Parameters:
maxResults(optional): Maximum emails to list (default: 10)label(optional): Filter by label (e.g., "INBOX", "UNREAD")
Getting Google OAuth Credentials
- Go to Google Cloud Console
- Create a new project or select existing
- Enable the Gmail API
- Create OAuth 2.0 credentials
- Use the OAuth playground or your app to get a refresh token
Response Format
All tools return structured JSON:
{
"success": true,
"messageId": "...",
"message": "Email sent successfully"
}
Part of Strider Labs
Strider Labs builds action execution infrastructure for AI agents.
- Website: https://striderlabs.ai
- GitHub: https://github.com/striderlabs
License
MIT
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.