A powerful Model Context Protocol (MCP) server providing comprehensive Gmail integration with LLM processing capabilities.
Configuration
View docs{
"mcpServers": {
"cablate-mcp-google-gmail": {
"command": "npx",
"args": [
"-y",
"@cablate/mcp-gmail"
],
"env": {
"GMAIL_CLIENT_ID": "your_client_id",
"GMAIL_CLIENT_SECRET": "your_client_secret",
"GMAIL_REFRESH_TOKEN": "your_refresh_token"
}
}
}
}This Gmail MCP Server provides seamless Gmail integration for your MCP environment, enabling AI-powered workflows over email with capabilities to read, search, filter, and process messages and attachments. It is designed to run alongside your existing MCP client, letting you invoke Gmail-backed actions from your language model workflows.
You can run the Gmail MCP Server from the command line using the provided CLI tool or add it to your Dive Desktop setup for a graphical, server-based integration.
CLI usage: start by invoking the Gmail MCP CLI. This launches the server utilities that bridge Gmail data into your MCP environment, enabling your models to read and process emails. Use it as part of your automation pipelines or within your conversational agents.
With Dive Desktop: add a new MCP server and paste the following configuration to integrate Gmail as a local MCP server.
{
"mcpServers": {
"gmail": {
"command": "npx",
"args": ["-y", "@cablate/mcp-gmail"],
"env": {
"GMAIL_CLIENT_ID": "your_client_id",
"GMAIL_CLIENT_SECRET": "your_client_secret",
"GMAIL_REFRESH_TOKEN": "your_refresh_token"
},
"enabled": true
}
}
}Prerequisites: you need Node.js and npm installed on your system. Node provides the runtime for the MCP server and the CLI tooling.
Step 1: Install the Gmail MCP server globally so you can access the CLI tools.
npm install -g @cablate/mcp-gmailStep 2: Verify the CLI is available and run the Gmail MCP command to start the server locally.
map-gmailTo enable Gmail access, you must configure Gmail API credentials (client ID, client secret, and a refresh token). Follow the Gmail API setup steps to generate these credentials, then supply them in the server configuration to authorize access to your Gmail account.
This server focuses on email-oriented actions within your MCP workspace. Use it to extract and route information from messages, search your mailbox, filter results, and handle attachments as part of your AI-driven workflows.
License: MIT. If you have questions or suggestions, you can reach out through the provided contact channels.
Read and search emails within Gmail, enabling you to retrieve message data based on queries or filters.
Process email content in various formats (text, HTML, attachments) for downstream AI processing.
Apply advanced filtering criteria to emails to narrow results for models and workflows.
Manage email attachments, including extraction, processing, and routing to storage or downstream tasks.