This is an experimental TMCP server that allows you to send emails via Gmail for demo purposes.
The client secret is stored and retrieved within the TSP Askar wallet. To set the client secret, use the following command with the contents of the client secret JSON file downloaded from the Google Cloud console:
tsp secret add gmail '{"installed":{"client_id":"...","project_id":"...","auth_uri":"...","token_uri":"...","auth_provider_x509_cert_url":"...","client_secret":"...","redirect_uris":["..."]}}'
After setting your client secret, run the Gmail server with:
uv run gmail_server.py
The server will then ask you to log in with the Gmail account you want to send emails from.
A powerful and flexible Gmail integration server built using the MCP (Message Control Protocol) framework. This server provides a robust interface to interact with Gmail APIs, offering functionality for reading, sending, and managing emails programmatically.
- Read emails from multiple Gmail accounts
- Send emails with attachments
- Search emails with advanced query options
- Download email attachments
- Handle email conversations and threads
- Real-time email monitoring
- Support for multiple Gmail accounts
Before running the Gmail MCP server, ensure you have the following:
- Python 3.12 or higher
- Google Cloud Project with Gmail API enabled
- OAuth 2.0 Client ID credentials
- Required Python packages (specified in pyproject.toml)
To install Gmail Integration Server for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @Quantum-369/Gmail-mcp-server --client claude
- Clone the repository:
git clone <your-repository-url>
cd gmail-mcp-server
- Create and activate a virtual environment:
python -m venv venv
# On Windows
venv\Scripts\activate
# On Unix/MacOS
source venv/bin/activate
- Install dependencies:
pip install .
- Install Python 3.12 and create a virtual environment if you haven't already:
sudo apt install python3.12 python3.12-venv -y python3.12 -m venv venv source venv/bin/activate
- Place your downloaded
client_secret.json
in the project root. - Generate an OAuth token for the Gmail account you want to use. Copy the URL
printed by the script into a web browser and complete the sign-in process:
python gmail_token_creator.py
- Start the server:
python gmail_server.py
- Go to the Google Cloud Console
- Create a new project or select an existing one
- Enable the Gmail API for your project
- Create OAuth 2.0 credentials:
- Go to "APIs & Services" > "Credentials"
- Click "Create Credentials" > "OAuth client ID"
- Choose "Desktop app" as application type
- Download the client configuration file
- Rename the downloaded file to
client_secret.json
and place it in the project root directory
- Set up email identifiers in
gmail_token_creator.py
:
email_identifier = '[email protected]' # Change this for each account
- Run the token creator to authenticate your Gmail accounts:
python gmail_token_creator.py
The script prints an authorization URL. Copy this URL into your web browser,
complete the Google consent flow, and copy the verification code back into the
terminal if prompted. A token file will be created inside token_files/
for
future use.
- Repeat the process for each Gmail account you want to integrate
gmail_server.py
: Main MCP server implementationgmail_api.py
: Gmail API interaction functionsgoogle_apis.py
: Google API authentication utilities- Supporting files:
read_emails.py
: Email reading functionalitysearch_emails.py
: Email search functionalitysend_emails.py
: Email sending functionality
python gmail_server.py
You can also build a container image using the provided Dockerfile
:
docker build -t gmail-mcp-server .
docker run -v $(pwd)/client_secret.json:/app/client_secret.json \
-v $(pwd)/token_files:/app/token_files gmail-mcp-server
The container runs the same server and stores authentication tokens in the
token_files
directory on the host so they persist between runs.
- Send Email:
await send_gmail(
email_identifier="[email protected]",
to="[email protected]",
subject="Test Subject",
body="Email body content",
attachment_paths=["path/to/attachment"]
)
- Search Emails:
await search_email_tool(
email_identifier="[email protected]",
query="from:[email protected]",
max_results=30,
include_conversations=True
)
- Read Latest Emails:
await read_latest_emails(
email_identifier="[email protected]",
max_results=5,
download_attachments=False
)
- Download Attachments:
await download_email_attachments(
email_identifier="[email protected]",
msg_id="message_id",
download_all_in_thread=False
)
- Store
client_secret.json
securely and never commit it to version control - Keep token files secure and add them to
.gitignore
- Use environment variables for sensitive information
- Regularly rotate OAuth credentials
- Monitor API usage and set appropriate quotas
The server includes comprehensive error handling and logging:
- Logs are written to
gmail_mcp.log
- Both file and console logging are enabled
- Detailed error messages for debugging
- Fork the repository
- Create a feature branch
- Commit your changes
- Push to the branch
- Create a Pull Request
Apachelicense2.0
For issues and feature requests, please use the GitHub issue tracker.