This project consists of two separate NestJS servers:
- Auth Server - Handles Google OAuth2 authentication
- Drive Server - Handles Google Drive CRUD operations
- Node.js (v14 or higher)
- npm or yarn
- Google Cloud Platform account with Google Drive API enabled
- OAuth 2.0 credentials from Google Cloud Console
-
Clone the repository
-
Install dependencies:
npm install
-
Create a
.envfile in the root directory with the following variables:# OAuth Server Configuration AUTH_SERVER_PORT=3000 AUTH_SERVER_URL=http://localhost:3000 # Drive Server Configuration DRIVE_SERVER_PORT=3001 DRIVE_SERVER_URL=http://localhost:3001 # Google OAuth Configuration GOOGLE_CLIENT_ID=your_client_id_here GOOGLE_CLIENT_SECRET=your_client_secret_here GOOGLE_REDIRECT_URI=http://localhost:3000/auth/google/callback # Session Configuration SESSION_SECRET=your_session_secret_here
-
Start the Auth Server:
npm run start:auth
-
Start the Drive Server:
npm run start:drive
GET /auth/google- Get Google OAuth URLGET /auth/google/callback- OAuth callback endpointGET /auth/tokens- Get current user's tokensPOST /auth/logout- Logout endpoint
GET /drive/files- List files- Query params:
pageSize,pageToken
- Query params:
GET /drive/files/:fileId- Get file detailsPOST /drive/files- Create new file- Body:
file(multipart/form-data),metadata(JSON)
- Body:
PUT /drive/files/:fileId- Update file- Body:
file(multipart/form-data),metadata(JSON)
- Body:
DELETE /drive/files/:fileId- Delete file
For development with hot-reload:
# Auth Server
npm run dev:auth
# Drive Server
npm run dev:drive- Always use HTTPS in production
- Implement proper session management
- Store tokens securely
- Implement rate limiting
- Add proper error handling
- Add request validation
- Implement proper logging
This MCP server allows Claude to access and interact with your Google Drive files. When properly configured, Claude can search, list, and read your files directly.
First, build the MCP server:
npm run build:mcpAdd the following configuration to your Claude Desktop configuration file:
{
"mcpServers": {
"google-drive-mcp": {
"command": "node",
"args": ["/path/to/google-drive-mcp/dist/index.js"],
"env": {
"GDRIVE_CREDS_DIR": "/path/to/config/directory",
"CLIENT_ID": "your-google-client-id",
"CLIENT_SECRET": "your-google-client-secret"
},
"description": "This server connects Claude to Google Drive, allowing it to search, list, and read your files. You can ask Claude to find specific documents, read file contents, or work with data from your Google Drive."
}
}
}This project is optimized for use with Claude Code, Anthropic's CLI tool. The combination enables powerful Google Drive automation and organization workflows.
- Bulk file operations: Delete empty folders, organize files, batch processing
- Intelligent categorization: Automated folder sorting using naming patterns
- Parallel processing: Efficient handling of large datasets with background jobs
- Token management: Optimized API usage to conserve Claude's context window
-
Mass File Organization:
# Claude Code can help organize hundreds of files into categories # Using parallel API calls and intelligent categorization
-
Empty Folder Cleanup:
# Automatically identify and remove empty folders # With verification and batch processing
-
Folder Structure Analysis:
# Analyze folder hierarchies and recommend optimizations # Generate reports on file distribution and usage patterns
The CLAUDE.md file contains detailed best practices for:
- Token conservation strategies
- Memory management techniques
- Parallel processing patterns
- API efficiency optimization
These patterns were developed through real-world usage organizing 300+ folders and are designed to minimize context usage while maximizing automation capabilities.