eyaltoledano/gdrive-mcp-server

Google Drive MCP Server

A powerful Model Context Protocol (MCP) server that provides seamless integration with Google Drive, allowing AI models to search, list, and read files from Google Drive.

🚀 Features

Tools

1. gdrive_search

Search for files in your Google Drive with powerful full-text search capabilities.

• Input:

  {
    "query": "string (your search query)"
  }
  

• Output: List of files with:
  • File name
  • MIME type
  • File ID
  • Last modified time
  • File size

2. gdrive_read_file

Read file contents directly using a Google Drive file ID.

• Input:

  {
    "file_id": "string (Google Drive file ID)"
  }
  

• Output: File contents with appropriate format conversion

Automatic File Format Handling

The server intelligently handles different Google Workspace file types:

• 📝 Google Docs → Markdown
• 📊 Google Sheets → CSV
• 📊 Google Presentations → Plain text
• 🎨 Google Drawings → PNG
• 📄 Text/JSON files → UTF-8 text
• 📦 Other files → Base64 encoded

🛠️ Getting Started

Prerequisites

• Node.js (v16 or higher)
• npm or yarn
• A Google Cloud Project
• A Google Workspace or personal Google account

Detailed Google Cloud Setup

1. Create a Google Cloud Project

   • Visit the Google Cloud Console
   • Click "New Project"
   • Enter a project name (e.g., "MCP GDrive Server")
   • Click "Create"
   • Wait for the project to be created and select it

2. Enable the Google Drive API

   • Go to the API Library
   • Search for "Google Drive API"
   • Click on "Google Drive API"
   • Click "Enable"
   • Wait for the API to be enabled

3. Configure OAuth Consent Screen

   • Navigate to OAuth consent screen
   • Select User Type:
     • "Internal" if you're using Google Workspace
     • "External" for personal Google accounts
   • Click "Create"
   • Fill in the required fields:
     • App name: "MCP GDrive Server"
     • User support email: your email
     • Developer contact email: your email
   • Click "Save and Continue"
   • On the "Scopes" page:
     • Click "Add or Remove Scopes"
     • Add https://www.googleapis.com/auth/drive.readonly
     • Click "Update"
   • Click "Save and Continue"
   • Review the summary and click "Back to Dashboard"

4. Create OAuth Client ID

   • Go to Credentials
   • Click "Create Credentials" at the top
   • Select "OAuth client ID"
   • Choose Application type: "Desktop app"
   • Name: "MCP GDrive Server Desktop Client"
   • Click "Create"
   • In the popup:
     • Click "Download JSON"
     • Save the file
     • Click "OK"

5. Set Up Credentials in Project

   # Create credentials directory
   mkdir credentials
   
   # Move and rename the downloaded JSON file
   mv path/to/downloaded/client_secret_*.json credentials/gcp-oauth.keys.json
   

Installation

# Clone the repository
git clone https://github.com/felores/gdrive-mcp-server.git
cd gdrive-mcp-server

# Install dependencies
npm install

# Build the project
npm run build

Authentication

1. Create a credentials directory and place your OAuth keys:

   mkdir credentials
   # Move your downloaded OAuth JSON file to the credentials directory as gcp-oauth.keys.json
   

2. Run the authentication command:

   node dist/index.js auth
   

3. Complete the OAuth flow in your browser

4. Credentials will be saved in credentials/.gdrive-server-credentials.json

🔧 Usage

As a Command Line Tool

# Start the server
node dist/index.js

Integration with Desktop App

Add this configuration to your app's server settings:

{
  "mcpServers": {
    "gdrive": {
      "command": "node",
      "args": ["path/to/gdrive-mcp-server/dist/index.js"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "path/to/gdrive-mcp-server/credentials/gcp-oauth.keys.json",
        "MCP_GDRIVE_CREDENTIALS": "path/to/gdrive-mcp-server/credentials/.gdrive-server-credentials.json"
      }
    }
  }
}

Replace path/to/gdrive-mcp-server with the actual path to your installation directory.

Example Usage

1. Search for files:

   // Search for documents containing "quarterly report"
   const result = await gdrive_search({ query: "quarterly report" });
   

2. Read file contents:

   // Read a specific file using its ID
   const contents = await gdrive_read_file({ file_id: "your-file-id" });
   

🔒 Security

• All sensitive credentials are stored in the credentials directory
• OAuth credentials and tokens are excluded from version control
• Read-only access to Google Drive
• Secure OAuth 2.0 authentication flow

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📝 License

This MCP server is licensed under the MIT License. See the LICENSE file for details.

🔍 Troubleshooting

If you encounter issues:

1. Verify your Google Cloud Project setup
2. Ensure all required OAuth scopes are enabled
3. Check that credentials are properly placed in the credentials directory
4. Verify file permissions and access rights in Google Drive

📚 Additional Resources

• Google Drive API Documentation
• OAuth 2.0 for Desktop Apps
• Model Context Protocol Documentation