mcp-googletasks
This Model Context Protocol (MCP) server provides a bridge between LLMs and Google Tasks, allowing you to manage your task lists and tasks directly through Claude.
claude mcp add --transport stdio arpitbatra123-mcp-googletasks node /path/to/google-tasks-mcp/build/index.js \ --env GOOGLE_CLIENT_ID="your_client_id_here" \ --env GOOGLE_REDIRECT_URI="http://localhost:3000/oauth2callback" \ --env GOOGLE_CLIENT_SECRET="your_client_secret_here"
How to use
This MCP server provides a bridge between Claude and Google Tasks, exposing a set of tools to manage task lists and tasks from within Claude. The available actions include creating, listing, updating, and deleting task lists as well as creating, listing, updating, deleting, and completing individual tasks. There is also support for moving tasks and clearing completed tasks within a list. To authorize access, you first authenticate with Google through the built-in authentication flow. The server exposes an authenticate tool to generate an authorization URL, which you visit to grant access to your Google Tasks. After completing the sign-in, you’ll receive an authorization code that you input back into Claude using the set-auth-code tool to finalize authentication. Once authenticated, you can start issuing the standard MCP commands to interact with your Google Tasks data.
Key workflow tips:
- Use authenticate to start the OAuth flow and then set-auth-code with the code you receive in the browser.
- The server persists credentials to disk (typically under ~/.config/google-tasks-mcp/credentials.json) so you don’t need to re-authenticate after restarts.
- You can list and inspect task lists, then operate on specific lists by passing their IDs to the corresponding task-related commands (e.g., list-tasks, create-task, etc.).
- The complete-task and clear-completed-tasks options help optimize task management by finalizing work and tidying lists.
How to install
Prerequisites:
- Node.js 20+ (as indicated by the project’s engines)
- npm
- A Google Cloud Project with the Google Tasks API enabled
Installation steps:
- Clone the repository or download the project files.
- Install dependencies:
npm install
- Build the server:
npm run build
- Configure environment variables in your runtime environment or a local start script (see mcp_config example for required vars):
- GOOGLE_CLIENT_ID: your Google OAuth Client ID
- GOOGLE_CLIENT_SECRET: your Google OAuth Client Secret
- GOOGLE_REDIRECT_URI: optional; defaults to http://localhost:3000/oauth2callback if not provided
- Start the server via your preferred runtime (node) as defined in mcp_config:
# Example start, replace with your actual path if needed
node /path/to/google-tasks-mcp/build/index.js
- If integrating with Claude for Desktop, ensure Claude is configured to point to the same server and environment variables are accessible to the process.
Additional notes
Notes and tips:
- Ensure GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET are present at startup; the server will fail fast with clear error messages if they are missing or invalid.
- The GOOGLE_REDIRECT_URI is optional but recommended to align with your OAuth client settings; default is http://localhost:3000/oauth2callback.
- Credentials are stored securely on disk (permissions restricted), enabling token refresh without repeated sign-ins.
- If you encounter token expiration issues, verify that the credentials.json file exists and has the correct permissions (600).
- The build/index.js path in mcp_config should reflect your actual deployment path after running npm run build.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud