Skip to main content
Glama

Google Drive MCP Server

by ducla5
README.md2.22 kB
# Google Drive MCP Server A Model Context Protocol (MCP) server that provides AI agents with efficient access to Google Drive documents. The server downloads files locally, converts them to markdown, and offers selective content extraction to optimize context window usage. ## Features - **Document Processing**: Convert Google Docs, PDFs, and Word documents to markdown - **Selective Content Extraction**: Extract specific sections, page ranges, or keyword-based content - **Intelligent Caching**: Local caching with LRU eviction and TTL-based expiration - **Advanced Search**: Search files by name, type, content, and metadata - **Permission Management**: Respect Google Drive permissions and sharing settings - **MCP Protocol Compliance**: Full integration with MCP-compatible AI agents ## Project Structure ``` src/ ├── auth/ # Authentication management ├── cache/ # Caching system ├── content/ # Content processing pipeline ├── search/ # Search functionality ├── types/ # TypeScript interfaces and types └── index.ts # Main server entry point ``` ## Installation ```bash npm install ``` ## Configuration The server requires Google Drive API credentials. Create a `config.json` file or set environment variables: ### Environment Variables ```bash GOOGLE_CLIENT_ID=your_client_id GOOGLE_CLIENT_SECRET=your_client_secret GOOGLE_REDIRECT_URI=http://localhost:8080/callback CACHE_DIR=./cache CACHE_MAX_SIZE=1GB CACHE_TTL=3600 LOG_LEVEL=info MAX_FILE_SIZE=100MB ``` ## Development ```bash # Build the project npm run build # Start in development mode npm run dev # Type checking npm run type-check # Linting npm run lint ``` ## MCP Tools The server exposes the following MCP tools: - `drive_search_files`: Search files on Google Drive - `drive_get_file`: Get file information and content - `drive_get_content_chunk`: Extract specific content portions - `drive_get_file_metadata`: Retrieve file metadata and permissions - `drive_list_folder`: List folder contents - `drive_get_comments`: Get file comments (if available) ## Requirements - Node.js >= 18.0.0 - Google Drive API credentials - TypeScript 5.5+ ## License MIT

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ducla5/gdriver-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server