google-chat

This MCP server adds a Google Chat integration to your MCP tooling. It exposes a Google Chat tools suite that lets you list available Google Chat spaces and fetch messages from specific spaces with optional date filtering. The server operates in conjunction with a separate authentication server that handles Google OAuth2 tokens, enabling secure access to your Google Chat data. To use it, first configure OAuth2 in Google Cloud and obtain the credentials JSON, then start the MCP server via the uv tool as described in the configuration instructions. You can then invoke the provided tools (get_chat_spaces and get_space_messages) through your MCP client to retrieve spaces and messages programmatically.

Prerequisites:

• Python 3.13+
• Google Cloud project with Google Chat API and People API enabled
• OAuth2 credentials downloaded as credentials.json placed in the project root

Installation steps:

1. Clone the repository git clone https://github.com/chy168/google-chat-mcp-server.git cd google-chat-mcp-server

2. Set up a Python environment (recommended) python -m venv venv source venv/bin/activate

3. Install required dependencies pip install -r requirements.txt (If a requirements file is not present, ensure google-api-python-client and google-auth-related packages are installed.)

4. Prepare token and credentials

   • Place credentials.json in the project root as referenced by your OAuth flow
   • The first run will generate a token.json after completing OAuth2 authorization

5. Run the authentication server or proceed with the CLI/web options described in the README to obtain a token For CLI mode: uv run python server.py --auth cli For Web mode: uv run python server.py --auth web --port 8000

6. Start the MCP server using the provided mcp.json configuration (see mcp_config). For example: uv --directory <YOUR_REPO_PATH>/google-chat-mcp-server run server.py --token-path <YOUR_REPO_PATH>/google-chat-mcp-server/token.json

Notes:

• Ensure the token.json path in the command matches where you store your tokens so the MCP server can access Google APIs.
• The Google Cloud OAuth flow requires you to configure redirect URIs (http://localhost:8000/auth for web mode) and download the credentials.json file into the project root.
• If running in containers, mount your project directory containing token.json and credentials.json into the container and pass --token-path accordingly.
• The Tools section exposes get_chat_spaces() and get_space_messages(space_name, start_date, end_date=None); use these to programmatically fetch data for automation or dashboards.
• If you encounter authentication errors, verify that the OAuth client is configured for the correct redirect URI and that the token.json has not expired. Regenerate tokens as needed via the authentication flow.