MCP Gmail Server

Provides MCP access to read Gmail messages and draft new emails using model context protocol.

other

0

GitHub Stars

other

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "vatskan-mcp-gmail-server": {
      "command": "uv",
      "args": [
        "run",
        "-m",
        "src.mcp_server_gmail"
      ],
      "env": {
        "PYTHONPATH": "/path/to/your/project/src"
      }
    }
  }
}

You set up a Gmail MCP Server that uses the model context protocol to read emails and draft replies in Gmail. It runs locally so you can test and automate email workflows with an MCP client.

How to use

Connect with your MCP client and start issuing tasks to the Gmail MCP Server. You can fetch recent emails, read unread messages, and generate draft emails based on your prompts. Practical usage patterns include getting the last few unread messages and creating draft replies for those messages.

How to install

Prerequisites: ensure you have a Python runtime and the UV core tool installed on your system.

Install UV (the asynchronous runner) on your system.

# Install UV (example command; follow the platform-specific installation guide)
# If you already have UV installed, you can skip this step

# No code snippet needed beyond the uv installation step; see your platform's docs for exact commands.

Install dependencies and prepare the environment using UV.

uv sync

Start the Gmail MCP Server with UV pointing to the Gmail MCP module.

uv run -m src.mcp_server_gmail

Additional sections

Configuration and runtime details are shown here so you can reproduce and customize the server in your environment.

OAuth client secrets for Google Gmail access are required. Create a Google Cloud project, enable the Gmail API, and configure OAuth 2.0 credentials. Download the client secrets as a JSON file and save it in your project root as client_secret_oauth_gcp.json (store securely and do not commit to version control). Refer to the setup steps for the exact sequence of actions: creating the project, enabling the Gmail API, configuring OAuth consent, and generating the OAuth client ID for a Desktop app.

Claude Desktop integration can run the Gmail MCP Server by launching the Python process with the appropriate module and environment settings. The example shows how to configure the command, arguments, and environment for the local server.

Test usage via Claude Desktop by asking to fetch unread emails and draft replies to them. You can try: “Get last 3 unread emails” and then “Create draft emails for these.”