MyGoogleDriveServer

jitzian/MyGoogleDriveServer

3.2

If you are the rightful owner of MyGoogleDriveServer and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

This is an MCP server that provides access to Google Drive files through various tools.

Tools
2
Resources
0
Prompts
0

Google Drive MCP Server

This is an MCP (Model Context Protocol) server that provides access to Google Drive files through two main tools:

Features

  1. List Drive Files - Lists all files in the Google Drive with metadata
  2. Read File Content - Reads and processes content from various file types including:
    • Text files (.txt, .md, .json, etc.)
    • Documents (.doc, .docx, .pdf)
    • Images (.png, .jpg, .gif, etc.) - with metadata extraction - TO BE IMPLEMENTED
    • Videos (.mp4, .mov, etc.) - with metadata extraction - TO BE IMPLEMENTED
    • Spreadsheets (.xls, .xlsx)
    • Presentations (.ppt, .pptx) - TO BE IMPLEMENTED

Setup

Prerequisites

Before you can use this MCP server, you need to set up a Google Cloud Platform service account and enable the Google Drive API.

Step 1: Create a Google Cloud Platform Project

  1. Sign in to Google Cloud Console: Navigate to Google Cloud Console and sign in with your Google account.
  2. Create a New Project:
    • Click on the project dropdown at the top of the page
    • Click "New Project"
    • Enter a project name (e.g., "MCP-Drive-Server")
    • Click "Create"

Reference: Create a Google Cloud Platform service account

Step 2: Enable Google Drive API

  1. Navigate to API Library:
    • In the Google Cloud Console, click on "APIs & Services" > "Library"
    • Use the left navigation menu or search for it
  2. Enable Google Drive API:
    • Search for "Google Drive API" in the search bar
    • Click on "Google Drive API" from the results
    • Click the "Enable" button

Reference: Using Google Drive API with Python and a service account

Step 3: Create a Service Account

  1. Navigate to Service Accounts:
    • Click on "IAM & Admin" > "Service Accounts" from the left navigation menu
  2. Create Service Account:
    • Click "Create Service Account" button at the top
    • Enter a service account name (e.g., "mcp-drive-server")
    • Optionally add a description
    • Click "Create and Continue"
  3. Grant Permissions (Optional):
    • You can skip this step for basic Drive access
    • Click "Continue" and then "Done"

Important: Service accounts are used for machine-to-machine authentication and don't require username/password. They're ideal for automated processes.

Step 4: Generate and Download JSON Key

  1. Access Service Account Keys:
    • In the Service Accounts list, find your newly created service account
    • Click on the service account email to open its details
    • Navigate to the "Keys" tab
  2. Create New Key:
    • Click "Add Key" > "Create new key"
    • Select "JSON" as the key type
    • Click "Create"
  3. Download and Store Securely:
    • A JSON file will be automatically downloaded to your computer
    • IMPORTANT: Never commit this file to version control (Git, GitHub, etc.)
    • Store it securely as it contains sensitive credentials

Reference: Step-by-Step: How to Create a Service Account to Access Google API

Step 5: Configure the MCP Server

  1. Place the JSON Key File:

    • Rename your downloaded JSON file to service-account.json
    • Move it to src/main/resources/service-account.json in this project
    • The file should be in the resources folder so it's included in the classpath

  2. Share Google Drive Files with Service Account:

    • Open your service-account.json file
    • Find the client_email field (e.g., mcp-drive-server@your-project.iam.gserviceaccount.com)
    • In Google Drive, share the folders/files you want to access with this email address
    • Give it "Viewer" or "Editor" permissions as needed

Step 6: Build and Run the Server

# Build the project
./gradlew build

# Run the server
./gradlew run

The server will start on http://localhost:3002/sse

Alternative: Using the Management Script

You can also use the provided management script:

# Start the server
./manage_server.sh start

# Stop the server
./manage_server.sh stop

# Check server status
./manage_server.sh status

Architecture

The codebase follows clean architecture principles:

src/main/kotlin/
├── Main.kt                    # Application entry point
├── GoogleDriveService.kt      # Google Drive API client
├── constants/
│   └── GlobalConstants.kt     # Application constants
├── domain/
│   ├── FileInfo.kt           # File metadata model
│   └── FileContent.kt        # File content model
└── tools/                    # MCP Tools (following single responsibility)
    ├── ToolManager.kt        # Registers all tools
    ├── ListDriveFilesTool.kt # Lists Google Drive files
    ├── ReadFileContentTool.kt # Reads file contents
    └── FileContentExtractor.kt # Extracts content from various file formats

Available Tools

1. list_drive_files

Lists all files in the Google Drive root folder.

Request:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "list_drive_files",
    "arguments": {}
  }
}

Response: Returns a JSON array of FileInfo objects with:

  • id: Google Drive file ID
  • name: File name
  • mimeType: MIME type of the file
  • size: File size in bytes
  • modifiedTime: Last modification timestamp

2. read_file_content

Reads and processes the content of a specific file.

Request:

{
  "jsonrpc": "2.0",
  "id": 2,
  "method": "tools/call",
  "params": {
    "name": "read_file_content",
    "arguments": {
      "fileId": "your-google-drive-file-id"
    }
  }
}

Response: Returns a FileContent object with:

  • id: File ID
  • name: File name
  • mimeType: MIME type
  • size: File size
  • content: Extracted content (text, metadata description, etc.)
  • contentType: Type of content extraction performed
  • metadata: Additional metadata extracted from the file

Content Processing Capabilities

The server can handle various file types:

  • Text Files: Direct text extraction
  • Documents (DOC, DOCX): Text extraction using Apache Tika
  • PDFs: Text extraction using Apache Tika
  • Images: Metadata extraction (dimensions, format, EXIF data)
  • Videos: Metadata extraction (format, duration, codec info)
  • Spreadsheets: Content extraction from Excel files
  • Binary Files: Basic metadata and description

Dependencies

Key libraries used:

  • Kotlin MCP SDK: For MCP protocol implementation
  • Google Drive API: For accessing Google Drive
  • Apache Tika: For content extraction from various file formats
  • Apache POI: For Microsoft Office document processing
  • PDFBox: For PDF processing
  • Ktor: For HTTP server functionality

Usage with MCP Clients

This server is designed to work with MCP-compatible clients like Claude Desktop or other AI assistants that support the Model Context Protocol.

To connect this server to Claude Desktop, add it to your MCP configuration:

{
  "mcpServers": {
    "google-drive": {
      "command": "java",
      "args": ["-jar", "path/to/your/server.jar"],
      "env": {}
    }
  }
}

Error Handling

The server includes comprehensive error handling for:

  • Google Drive API errors
  • File processing errors
  • Unsupported file types
  • Network connectivity issues

All errors are returned as proper MCP error responses with descriptive messages.