Google Drive MCP
Enables Claude to read, create, update, and manage files in Google Drive through MCP tools.
README
Google Drive MCP Server
Connect Google Drive to Claude Code with a simple double-click setup. This MCP server allows Claude to read, create, update, and manage files in your Google Drive.
Prerequisites
- Python 3.10 or higher
- A Google Account (personal or Google Workspace)
- Google Cloud Console access
Step 1: Create Google Cloud Credentials
1.1 Create a Google Cloud Project
- Go to Google Cloud Console
- Click the project dropdown at the top of the page
- Click New Project
- Enter a project name (e.g., "Google Drive MCP")
- Click Create
- Wait for the project to be created, then select it
1.2 Enable the Google Drive API
- In the Google Cloud Console, go to APIs & Services > Library
- Search for "Google Drive API"
- Click on Google Drive API
- Click Enable
1.3 Configure OAuth Consent Screen
- Go to APIs & Services > OAuth consent screen
- Select External (or Internal if using Google Workspace)
- Click Create
- Fill in the required fields:
- App name: Google Drive MCP
- User support email: Your email
- Developer contact email: Your email
- Click Save and Continue
- On the Scopes page, click Add or Remove Scopes
- Add these scopes:
https://www.googleapis.com/auth/drive.readonlyhttps://www.googleapis.com/auth/drive.filehttps://www.googleapis.com/auth/drive
- Click Update, then Save and Continue
- On the Test users page, click Add Users
- Add your email address
- Click Save and Continue
1.4 Create OAuth Client ID
- Go to APIs & Services > Credentials
- Click Create Credentials > OAuth client ID
- Select Desktop app as the application type
- Enter a name (e.g., "Google Drive MCP Desktop")
- Click Create
- Click Download JSON on the popup
- Rename the downloaded file to
credentials.json - Place
credentials.jsonin the project root directory
Important: The credentials must be of type "Desktop app", not "Web application".
Step 2: Install Dependencies
Option A: Using Virtual Environment (Recommended)
# Clone the repository
git clone https://github.com/david-valdivia/GoogleDriveMCP.git
cd GoogleDriveMCP
# Create virtual environment (Python 3.10+ required)
python3 -m venv .venv
# Activate virtual environment
# On macOS/Linux:
source .venv/bin/activate
# On Windows:
.venv\Scripts\activate
# Install dependencies
pip install -r requirements.txt
Option B: Global Installation
pip install -r requirements.txt
Step 3: Run the Setup GUI
# With virtual environment activated
python -m src.main
This opens the setup window where you can:
- Select your permission level (Read Only, Read + Create, or Full Access)
- Click "Connect to Google" to authenticate
- Complete the OAuth flow in your browser
- Click "Configure MCP" to select a project directory for Claude Code
Step 4: Build the Executable (Optional)
Build a standalone executable that doesn't require Python:
# With virtual environment activated
python build/build.py
Build Output
| Platform | Output |
|---|---|
| macOS | build/dist/Google Drive MCP.app and build/dist/googledrive-mcp |
| Windows | build/dist/googledrive-mcp.exe |
| Linux | build/dist/googledrive-mcp |
macOS Additional Requirements
If building on macOS, you need Tcl/Tk for the GUI:
# Install Homebrew if not already installed
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
# Install Python with Tcl/Tk support
brew install python-tk@3.12
Step 5: Configure Claude Code
After authentication, the app automatically creates a .mcp.json file in your selected project directory. Alternatively, create it manually:
{
"mcpServers": {
"googledrive": {
"command": "/path/to/googledrive-mcp",
"args": ["--serve"]
}
}
}
Replace /path/to/googledrive-mcp with:
- Built executable: Full path to
googledrive-mcporgoogledrive-mcp.exe - From source: Full path to Python and script, e.g.:
{ "mcpServers": { "googledrive": { "command": "/path/to/.venv/bin/python", "args": ["/path/to/GoogleDriveMCP/src/main.py", "--serve"] } } }
Permission Levels
| Level | Scopes | Capabilities |
|---|---|---|
| Read Only | drive.readonly |
View and search files |
| Read + Create | drive.file |
Above + create new files |
| Full Access | drive |
Read, create, update, delete, share |
Available MCP Tools
| Tool | Description | Required Permission |
|---|---|---|
list_files |
List files in a folder or root | Read Only |
read_file |
Read file contents (exports Google Docs to text) | Read Only |
search_files |
Full-text search across Drive | Read Only |
list_folders |
List all folders for navigation | Read Only |
get_sharing_info |
View who has access to a file | Read Only |
create_file |
Create a new file or Google Doc | Read + Create |
create_folder |
Create a new folder | Read + Create |
update_file |
Update file contents | Full Access |
delete_file |
Move file to trash | Full Access |
move_file |
Move file to different folder | Full Access |
copy_file |
Copy a file | Full Access |
share_file |
Share file with email address | Full Access |
Development
Run Tests
# With virtual environment activated
pytest -v
Project Structure
GoogleDriveMCP/
├── src/
│ ├── main.py # Entry point, mode detection
│ ├── gui.py # Setup GUI with tkinter
│ ├── server.py # MCP server implementation
│ ├── drive.py # Google Drive API wrapper
│ ├── auth.py # OAuth flow and token management
│ └── config.py # Platform-specific config paths
├── tests/ # Test files
├── build/
│ ├── build.py # Build script
│ └── googledrive-mcp.spec # PyInstaller spec
├── credentials.json # Your OAuth credentials (not committed)
├── requirements.txt # Python dependencies
└── README.md
Token Storage Location
Tokens are stored in platform-specific locations:
| Platform | Location |
|---|---|
| macOS | ~/Library/Application Support/GoogleDriveMCP/ |
| Windows | %APPDATA%\GoogleDriveMCP\ |
| Linux | ~/.config/GoogleDriveMCP/ |
Troubleshooting
"credentials.json not found"
Make sure credentials.json is in the project root directory (same folder as src/).
"Error 400: redirect_uri_mismatch"
Your OAuth credentials are configured as "Web application" instead of "Desktop app". Create new credentials with type "Desktop app".
"Access blocked: This app's request is invalid"
- Make sure you added yourself as a test user in the OAuth consent screen
- Verify the app is in "Testing" mode (not "In production")
"ModuleNotFoundError: No module named 'tkinter'"
On macOS with Homebrew Python:
brew install python-tk@3.12
On Ubuntu/Debian:
sudo apt-get install python3-tk
MCP server not connecting in Claude Code
- Verify the path in
.mcp.jsonis correct and absolute - Restart Claude Code after creating/modifying
.mcp.json - Check that the executable has execute permissions:
chmod +x googledrive-mcp
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.