Google Calendar
Integrates with Google Calendar to manage events and generate calendar insights.
Google Calendar MCP Server
Interact with your Google Calendar through Claude Desktop using the Model Context Protocol (MCP).
This is a TypeScript-based MCP server that implements Google Calendar integration. It demonstrates core MCP concepts while providing:
- Calendar event management through MCP URIs
- Tools for creating and modifying events
- Prompts for generating calendar insights
Features
Resources
- Access calendar events via MCP URIs
- Each event has title, time, description, and attendees
- Structured event data with proper mime types
Tools
create_event- Create new calendar events- Takes title, time, and other event details as parameters
- Directly interfaces with Google Calendar API
list_events- View upcoming calendar events- [Add other tools you've implemented]
Prompts
analyze_schedule- Generate insights about your calendar- Includes upcoming events as embedded resources
- Returns structured prompt for LLM analysis
- [Add other prompts you've implemented]
Prerequisites
- Node.js (v14 or higher)
- A Google Cloud Project with Calendar API enabled
- OAuth 2.0 Client credentials
Development
Install devbox by following instructions at devbox.sh
curl -fsSL https://get.jetpack.io/devbox | bash
Initialize devbox in the project directory:
devbox init
Start the devbox shell:
devbox shell
Install dependencies:
npm install
Build the server:
npm run build
For development with auto-rebuild:
npm run watch
Installation
To use with Claude Desktop, add the server config:
On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"Google Calendar": {
"command": "/path/to/Google Calendar/build/index.js"
}
}
}
First-Time Setup
-
Set up Google Cloud credentials:
- Go to Google Cloud Console
- Create a new project or select an existing one
- Enable the Google Calendar API
- Create OAuth 2.0 credentials (Desktop application type)
- Download the client secret JSON file
- Rename it to
.client_secret.jsonand place it in the project root
-
Initial Authentication:
- When first running the server, it will provide an authentication URL
- Visit the URL in your browser
- Grant the requested permissions
- Copy the provided authorization code
- Paste the code back into the CLI prompt
Debugging
Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:
npm run inspector
The Inspector will provide a URL to access debugging tools in your browser.
Related Servers
Atlassian
Interact with Atlassian tools like Confluence and Jira.
CData Jira Service Management
A read-only server to query live Jira Service Management data via a simple MCP interface, powered by CData.
Google Tasks
Integrate with Google Tasks to manage your to-do lists and tasks.
Evernote
Connects your Evernote account to an LLM, enabling natural language search and queries over your notes.
Retrieval Augmented Thinking
A server implementing Chain of Draft reasoning for enhanced problem-solving capabilities using OpenAI.
ClickUp
Interact with ClickUp's task management API to manage projects and tasks through natural language.
MCP Orchestrator
A universal interface to manage and interact with all your MCP servers from a single point, using external configuration files for mappings and credentials.
Desktop Automation
Control your desktop with AI. Automate mouse movements, keyboard inputs, and screen captures.
MCP Chatbot
An intelligent chatbot for automating tasks like browser control, web searches, and travel planning.
OmniTaskAgent
A multi-model agent for managing tasks across various platforms, requiring API keys for different AI models.