mcpField
G
Google Tasks MCP Server

Google Tasks MCP Server

Local
github.com/arpitbatra123Productivity51Updated 5w ago

About Google Tasks MCP Server

Google Tasks MCP Server is an MCP server that connects MCP clients to Google Tasks, letting users manage task lists and tasks from Claude Desktop, Cursor, or Codex. It is aimed at users who want to handle to-do items within AI conversations, and it stores OAuth credentials on disk so re-authentication is rarely needed.

Key features

  1. 01Create, read, update and delete task lists
  2. 02Manage tasks including completion status
  3. 03Move tasks within or across task lists
  4. 04Clear completed tasks from a list
  5. 05Persistent OAuth credentials with automatic token refresh

Use cases

  1. 01Organizing a daily to-do list from Claude Desktop
  2. 02Bulk updating or cleaning up task lists through AI chat
  3. 03Checking and completing tasks from Codex or Cursor

MCP support

This project is an MCP server that exposes tools such as list-tasklists, create-task and complete-task to MCP clients for operating on Google Tasks data.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

Google Tasks MCP Server

This Model Context Protocol (MCP) server provides a bridge between MCP clients and Google Tasks, allowing you to manage your task lists and tasks directly from clients like Claude Desktop, Cursor, and Codex.

Note

All (bar some edits) code in this project was "vibe coded" - generated with Claude/Copilot with instructions from me.

Features

This MCP server provides the following functionality:

Task List Management
  • list-tasklists - List all your task lists
  • get-tasklist - Get details about a specific task list
  • create-tasklist - Create a new task list
  • update-tasklist - Update an existing task list
  • delete-tasklist - Delete a task list
Task Management
  • list-tasks - List all tasks in a task list, including paginated results
  • get-task - Get details about a specific task
  • create-task - Create a new task
  • update-task - Update an existing task
  • delete-task - Delete a task
  • complete-task - Mark a task as completed
  • move-task - Move a task (reorder, change parent, or move across task lists)
  • clear-completed-tasks - Clear all completed tasks from a list

Setup Instructions

1. Create Google Cloud Project & Enable API
  1. Go to the Google Cloud Console
  2. Create a new project
  3. Navigate to "APIs & Services" > "Library"
  4. Search for "Google Tasks API" and enable it
  5. Go to "APIs & Services" > "Credentials"
  6. Click "Create Credentials" > "OAuth Client ID"
  7. Configure the OAuth consent screen (External is fine for personal use)
  8. For Application Type, select "Web application"
  9. Add "http://localhost:3000/oauth2callback" as an authorized redirect URI
  10. Create the client ID and secret
2. Configure Environment Variables

Create a .env file in this project directory:

GOOGLE_CLIENT_ID=your_client_id_here
GOOGLE_CLIENT_SECRET=your_client_secret_here
GOOGLE_REDIRECT_URI=http://localhost:3000/oauth2callback

Environment Variables:

  • GOOGLE_CLIENT_ID (required) - Your Google OAuth Client ID
  • GOOGLE_CLIENT_SECRET (required) - Your Google OAuth Client Secret
  • GOOGLE_REDIRECT_URI (optional) - OAuth redirect URI (defaults to http://localhost:3000/oauth2callback)

Note: The server validates that GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET are set at startup and will fail with clear error messages if they are missing or invalid.

3. Build the Server
  1. Install dependencies:
npm install
  1. Build the server:
npm run build
4. Configure an MCP Client
Codex

Add the server to ~/.codex/config.toml:

[mcp_servers.google-tasks]
command = "zsh"
args = ["-lc", "cd /path/to/google-tasks-mcp && exec node build/index.js"]

Replace /path/to/google-tasks-mcp with the path to this project, then restart Codex.

To verify the server is registered:

codex mcp list
codex mcp get google-tasks
Claude for Desktop
  1. Install Claude for Desktop
  2. Open the Claude configuration file:
    • MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
  3. Add the Google Tasks MCP server configuration:
{
  "mcpServers": {
    "google-tasks": {
      "command": "node",
      "args": ["/path/to/google-tasks-mcp/build/index.js"]
    }
  }
}

Replace the path with your own value, then restart Claude for Desktop.

Authentication Flow

When you first use the Google Tasks MCP server:

  1. Use the authenticate tool to get an authorization URL
  2. Visit the URL in your browser and sign in with your Google account
  3. After authorizing, you'll receive a code on the browser page
  4. Use the set-auth-code tool with this code to complete authentication
  5. You can now use all other tools to interact with Google Tasks

Note: Your authentication tokens (including refresh tokens) are automatically saved to disk at ~/.config/google-tasks-mcp/credentials.json with restricted permissions (600). This means:

  • You only need to authenticate once - credentials persist across server restarts
  • Automatic token refresh - Access tokens are automatically refreshed when they expire (typically after 1 hour) using the saved refresh token
  • No re-authentication needed - After the initial setup, you won't need to authenticate again unless you revoke access or delete the credentials file

Requirements

  • Node.js 20+ (see package.json engines)
  • An MCP-compatible client
  • Google Cloud Project with Tasks API enabled

Implementation Features

This MCP server includes the following improvements:

  • Persistent token storage - Authentication credentials are saved to disk (~/.config/google-tasks-mcp/credentials.json) with restricted permissions, so you only need to authenticate once
  • Environment variable validation - Startup validation ensures required credentials are configured with clear error messages
  • Automatic token refresh - OAuth tokens are automatically refreshed when they expire, eliminating the need to re-authenticate during active sessions or after restarts
  • Enhanced input validation - Comprehensive validation of all inputs including ID formats, string lengths, and RFC 3339 date formats
  • HTML sanitization - OAuth callback responses are sanitized to prevent XSS vulnerabilities
  • Graceful shutdown - Proper cleanup of resources on SIGINT/SIGTERM signals
  • Type safety - Full TypeScript type safety throughout the codebase with proper interfaces
  • Configurable redirect URI - The OAuth redirect URI can be customized via the GOOGLE_REDIRECT_URI environment variable

Screenshots

Claude Desktop

Claude Desktop Screenshot

Cursor

Cursor Screenshot

License

This project is for demonstration purposes only. Use at your own risk.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.