google-tasks-mcp
Enables AI assistants to manage Google Tasks, including listing, creating, updating, deleting, and completing tasks via the Google Tasks API.
README
Google Tasks MCP Server
A Model Context Protocol (MCP) server for Google Tasks. Lets AI assistants list, create, update, delete, and complete tasks via the Google Tasks API.
Why this exists
I prefer Google Tasks over Google Calendar for managing todos. The excellent google-calendar-mcp project only covers the Calendar API, so I built this quick-and-dirty companion to fill the gap. If you'd like to fold Tasks support into that project, PRs are welcome there.
Quick Start
1. Enable the Google Tasks API
- Go to the Google Cloud Console
- Create a new project or select an existing one
- Enable the Google Tasks API (make sure the correct project is selected in the top bar)
- Create OAuth 2.0 credentials:
- Go to Credentials
- Click Create Credentials > OAuth client ID
- Choose Desktop app as the application type
- Download the JSON file (you'll reference its path below)
- Configure the OAuth consent screen:
- Add your email as a Test User
- Note: test-mode tokens expire after 7 days (see Limitations)
2. Install
Add to your MCP client config (Cursor, Claude Desktop, etc.):
{
"mcpServers": {
"google-tasks": {
"command": "npx",
"args": ["-y", "@scottie-will/google-tasks-mcp"],
"env": {
"GOOGLE_OAUTH_CREDENTIALS": "/path/to/your/gcp-oauth.keys.json"
}
}
}
}
3. Authenticate
Option A — CLI (recommended for first run):
GOOGLE_OAUTH_CREDENTIALS=/path/to/gcp-oauth.keys.json npx @scottie-will/google-tasks-mcp auth
Complete the OAuth flow in your browser, then restart your MCP client.
Option B — Via the agent:
Ask your AI assistant to "authenticate with Google Tasks". The authenticate tool will return a URL to visit in your browser. After completing sign-in, tell the agent to verify authentication.
Available Tools
| Tool | Description |
|---|---|
authenticate |
Start or verify Google OAuth authentication |
list-task-lists |
List all task lists for the authenticated user |
list-tasks |
List tasks with optional date filtering and completion status |
get-task |
Get a specific task by ID |
create-task |
Create a new task (supports subtasks via parent) |
update-task |
Update an existing task (partial updates supported) |
delete-task |
Permanently delete a task |
complete-task |
Mark a task as completed |
Environment Variables
| Variable | Required | Description |
|---|---|---|
GOOGLE_OAUTH_CREDENTIALS |
Yes | Path to your OAuth credentials JSON file |
GOOGLE_TASKS_MCP_TOKEN_PATH |
No | Custom path for token storage (default: ~/.config/google-tasks-mcp/tokens.json) |
Limitations
- Dates only, no times. The Google Tasks API stores due dates only — any time component in the
duefield is silently discarded. Workaround: prefix the task title with a 24-hour time (e.g. "0900 Team standup") for manual time-based ordering. - Position-based ordering only. Tasks are returned in position order (the drag-order from the Google Tasks UI). There is no server-side sort by due date — sort client-side after fetching.
- No pagination. Each list call returns at most 100 items. Large task lists may be truncated.
- Test-mode token expiry. While your GCP app is in test mode, OAuth tokens expire after 7 days. To avoid this, publish your app to production (no verification needed for personal use — Google will show an "unverified app" warning).
- Single account. Only one Google account is supported at a time.
For full API details, see the Google Tasks API reference.
Agent Setup Instructions
This server includes built-in instructions metadata (via the MCP protocol) that teaches agents how to use it effectively — covering the auth flow, date-only behavior, sorting quirks, @default task list conventions, and destructive-operation guardrails. Compatible MCP clients (Cursor, Claude Desktop, etc.) surface these instructions automatically.
If your client supports serverUseInstructions in the MCP config, you can also add custom hints. See incident.io's MCP for an example of thorough server instructions.
GCP API Access Required
This server requires:
- API: Google Tasks API (
tasks.googleapis.com) - OAuth scope:
https://www.googleapis.com/auth/tasks(full read/write access to tasks) - Credential type: OAuth 2.0 Client ID — Desktop app type
No API key or service account is needed. The OAuth flow runs locally and tokens are stored on your machine at ~/.config/google-tasks-mcp/tokens.json.
Credits
- Inspired by and modeled after google-calendar-mcp by @nspady. That project covers the Google Calendar API with multi-account support, recurring events, and much more. If you'd like to add Tasks support there, open a PR — it would be a natural fit.
- Built with the Model Context Protocol SDK.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.