Gmail Server for Model Context Protocol (MCP)

Name: Gmail Server for Model Context Protocol (MCP)
Author: MCP-Mirror

@MCP-Mirror

リポジトリ

Gmail Server for Model Context Protocol (MCP) について

Mirror of

#gmail #mcp #email-server

基本情報

カテゴリ

コミュニケーション

ライセンス

GPL-3.0 license

ランタイム

python

トランスポート

stdio

公開者

MCP-Mirror

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "jasonsum_gmail-mcp-server": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "uv",
        "run",
        "[absolute-path-to-git-repo]/src/gmail/server.py",
        "--creds-file-path",
        "[absolute-path-to-credentials-file]",
        "--token-path",
        "[absolute-path-to-access-tokens-file]"
      ]
    }
  }
}

ツール

Sends email to email address recipient

Moves email to trash

Marks email as read

Retrieves unread emails

Retrieves given email content

Open email in browser

概要

What is Gmail Server for Model Context Protocol (MCP)?

This MCP server integrates with Gmail to enable sending, removing, reading, drafting, and responding to emails. The MCP client prompts the user before conducting such activities.

How to use Gmail Server for Model Context Protocol (MCP)?

First, set up a Google Cloud project, enable the Gmail API, configure an OAuth consent screen with test user, add OAuth scope https://www.googleapis.com/auth/gmail/modify, and create an OAuth Client ID for a Desktop App. Download the JSON credentials file and note its absolute path. Then configure an MCP client (e.g., Claude Desktop) using uv with parameters --creds-file-path and --token-path. On startup, a browser-based authentication flow saves token credentials to the specified token path.

Key features of Gmail Server for Model Context Protocol (MCP)

Send emails to a recipient with subject and content
Move emails to trash by ID
Mark emails as read by ID
Retrieve a list of unread emails
Fetch full content of a specific email (marks as read)
Open an email in the default browser by ID

Use cases of Gmail Server for Model Context Protocol (MCP)

An AI assistant sends an email on behalf of the user
Automatically trash or mark-as-read bulk messages
Read unread emails and summarize them for the user
Open a specific email in the browser for manual review

FAQ from Gmail Server for Model Context Protocol (MCP)

What Gmail OAuth scopes are required?

The server uses https://www.googleapis.com/auth/gmail.modify which allows reading, sending, and modifying emails.

How is authentication handled?

On server start, a browser-based OAuth flow launches. Token credentials are saved to the file specified by --token-path and reused on subsequent runs.

What are the required dependencies?

A Google Cloud project with the Gmail API enabled, a Desktop App OAuth client, and uv for running the server. No Python dependencies beyond those bundled with the server are mentioned.

Where are credentials and tokens stored?

Credentials are stored in the file path given to --creds-file-path. Access and refresh tokens are stored in the file path given to --token-path. Both are local file paths on the user’s machine.