How to use the Google Drive MCP server

Go to console.cloud.google.com and create a new project (or select an existing one). Navigate to APIs & Services > Library, search for 'Google Drive API', and click Enable. This allows your MCP server to make API calls to Google Drive on your behalf.

In your Google Cloud project, go to APIs & Services > Credentials. Click 'Create Credentials' > 'OAuth client ID'. Choose 'Desktop app' as the application type and give it a name like 'MCP Google Drive'. Download the JSON credentials file. You will need the client_id and client_secret from this file. If this is a new project, you may need to configure the OAuth consent screen first — choose 'External' user type and add your email as a test user.

1Steps:
21. Google Cloud Console > APIs & Services > Credentials
32. Create Credentials > OAuth client ID
43. Application type: Desktop app
54. Name: MCP Google Drive
65. Download JSON file
76. Extract client_id and client_secret values

Open your MCP host's configuration file and add a google-drive entry. Pass the client_id and client_secret as environment variables. The server will initiate an OAuth flow on first run, opening a browser window where you authorize access to your Google Drive.

1{
2  "mcpServers": {
3    "google-drive": {
4      "command": "npx",
5      "args": [
6        "-y",
7        "@anthropic-ai/google-drive-mcp"
8      ],
9      "env": {
10        "GOOGLE_CLIENT_ID": "your-client-id.apps.googleusercontent.com",
11        "GOOGLE_CLIENT_SECRET": "your-client-secret"
12      }
13    }
14  }
15}

Restart your AI host. On first launch, the Google Drive MCP server will open a browser window asking you to sign in with your Google account and authorize access to Google Drive. Grant the requested permissions. The server stores the refresh token locally so you only need to authenticate once. Future startups will use the stored token automatically.

Ask your AI to find and read files from your Drive. It can search by filename, content, file type, or folder. When reading Google Docs, the server extracts the text content. For Sheets, it can read cell data. For PDFs and other files, it returns available text content.

1Example prompts:
2
3"Search my Google Drive for files containing 'Q4 budget'"
4
5"Read the contents of the document called 'Product Requirements v2' in my Drive"
6
7"List all spreadsheets in my 'Project Alpha' folder"

For VS Code with GitHub Copilot, use the servers key instead of mcpServers. Create a .vscode/mcp.json file in your workspace root with the same command, args, and env values.

1{
2  "servers": {
3    "google-drive": {
4      "command": "npx",
5      "args": [
6        "-y",
7        "@anthropic-ai/google-drive-mcp"
8      ],
9      "env": {
10        "GOOGLE_CLIENT_ID": "your-client-id.apps.googleusercontent.com",
11        "GOOGLE_CLIENT_SECRET": "your-client-secret"
12      }
13    }
14  }
15}