Google Calendar MCP Server

🚀 Google Calendar MCP Server

This project provides a server that uses the Google Calendar API to search calendar content and configure it as an MCP server.

🚀 Quick Start

This document outlines the environmental configuration steps required to use the Google Calendar API to search calendar content and set it up as an MCP server.

✨ Features

• Utilize the Google Calendar API to search calendar content.
• Configure the server as an MCP server.

📦 Installation

Prerequisites

• uv must be installed.
• A Google account is required.

Steps

1. Create a Google Cloud Project

Enable the Google Calendar API in the Google Cloud Console or via the following link: Enable API

2. Configure the OAuth 2.0 Consent Screen

Navigate to the branding page in the Google Cloud Console menu or visit the following link: Branding

1. Set the application name.
2. Set the user support email.
3. Set the developer contact information.

3. Configure Desktop App Credentials

Navigate to the client page in the Google Cloud Console menu or visit the following link: Clients

1. Click Create credentials.
2. Select Desktop app under Application type.
3. Enter a name for the credentials in the Name field.
4. Click Create.
5. Download and save the generated JSON file as credentials.json, and move the file to your working directory.

4. Configure the Python Environment

Create a Virtual Environment

1. Open a terminal or command prompt.
2. Switch to the project directory.
3. Set up the virtual environment and download the required libraries.

uv sync

Generate token.json

1. Run the following command to generate the token.json file:

uv run generate_token.py

2. A browser will open, requesting permission to access your Google account. Grant the permission.
3. After verification is complete, the token.json file will be created in the project directory.

5. Configure calendar_settings.json

1. Edit the calendar_settings.json.sample file in the project directory. For example, configure it as follows:

{
  "calendar_ids": {
    "primary": "primary",
    "work": "example_work_schedule_id"
  }
}

• primary: The ID of your Google account's primary calendar (usually "primary").
• work: The ID of your work calendar. Set the required calendar IDs, then rename the file to calendar_settings.json and save it.

6. Configure MCP

1. Navigate to the .vscode directory in the project directory and edit the mcp.json.sample file.
2. Modify the mcp.json.sample file as follows:

{
  "servers": {
    "google-calendar-mcp-server": {
      "type": "stdio",
      "command": "/home/<username>/<dir>/.venv/bin/python",
      "args": [
        "/home/<username>/<dir>/google_calendar_mcp_server.py"
      ]
    }
  }
}

• Replace <username> with your username.
• Replace <dir> with the project directory name. For example, if your username is tatsuya and the project directory name is google-calendar-mcp-server, the configuration is as follows:

{
  "servers": {
    "google-calendar-mcp-server": {
      "type": "stdio",
      "command": "/home/tatsuya/google-calendar-mcp-server/.venv/bin/python",
      "args": [
        "/home/tatsuya/google-calendar-mcp-server/google_calendar_mcp_server.py"
      ]
    }
  }
}

3. After modification, rename the file to mcp.json and save it.

7. Configure Proxy Mode

Follow these steps to set up the proxy mode for GitHub Copilot Chat in Visual Studio Code:

1. Open Visual Studio Code and navigate to the Extensions view.
2. Search for and install the "GitHub Copilot Chat" extension.
3. After installation, restart Visual Studio Code.

8. Start the Proxy

1. Open the folder to be configured in Visual Studio Code.
2. Click the proxy icon (a robot avatar) in the right activity bar.
3. If it's your first time using it, you may be prompted to initialize the settings. Follow the prompts to complete the operation.
4. Enter the following command to start the proxy:

npm run dev

5. The proxy will run on the specified port, defaulting to localhost:3000.

Troubleshooting

• If an error occurs: Check the error message and reinstall the dependencies or check the settings in the Google Cloud Console as needed.

Chinese README is here