Google Drive MCP Server

What is Google Drive MCP Server?

A Model Context Protocol (MCP) server that integrates with Google Drive, enabling AI models to search, list, and read files from Google Drive. It is intended for developers building AI applications that need to access Google Drive content.

How to use Google Drive MCP Server?

Install dependencies, build the project, then set up Google Cloud credentials and run node dist/index.js auth to authenticate via OAuth. Start the server with node dist/index.js. For integration with a desktop MCP app, add a JSON configuration block specifying the command, arguments, and environment variables for the OAuth key files.

Key features of Google Drive MCP Server

• Two tools: gdrive_search (file search) and gdrive_read_file (read file contents by ID)
• Automatic format conversion for Google Workspace files (Docs to Markdown, Sheets to CSV, etc.)
• Read-only access to Google Drive via OAuth 2.0
• Credentials stored in a local credentials directory
• Works with Node.js v16 or higher and npm/yarn

Use cases of Google Drive MCP Server

• Search for files in Google Drive using full-text queries
• Read file contents directly by providing a Google Drive file ID
• Automatically convert Google Docs, Sheets, and Presentations into plain text formats for AI consumption
• Integrate Google Drive file access into MCP‑compatible desktop applications

FAQ from Google Drive MCP Server

What are the prerequisites for using this server?

You need Node.js v16 or higher, npm or yarn, a Google Cloud Project with the Google Drive API enabled, and a Google Workspace or personal Google account.

How do I authenticate with Google Drive?

After placing your downloaded OAuth client JSON file in credentials/gcp-oauth.keys.json, run node dist/index.js auth and complete the OAuth flow in your browser. Credentials are saved to credentials/.gdrive-server-credentials.json.

What permissions does the server request?

The server requests read‑only access to Google Drive (scope https://www.googleapis.com/auth/drive.readonly).

How are credentials stored and managed?

All sensitive credentials are stored in the credentials directory, which is excluded from version control. The OAuth keys and token files are kept locally.

What file types are supported and how are they converted?

Google Docs become Markdown, Google Sheets become CSV, Google Presentations become plain text, Google Drawings become PNG, text/JSON files are returned as UTF‑8 text, and other files are Base64 encoded.