Google Tasks MCP Server
Let your coding agent list, create, and complete Google Tasks without leaving the terminal or IDE.
Overview
Google Tasks MCP Server is a MCP server for the Build phase that connects MCP-compatible AI clients to Google Tasks via OAuth so agents can manage task lists programmatically.
What is this MCP server?
- Stdio MCP server (PyPI package google-tasks-mcp v0.4.0) for Google Tasks API operations from compatible agents
- Google OAuth via GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, and GOOGLE_REDIRECT_URI with token persistence in SQLite (DB_PA
- Optional DEFAULT_TASKLIST to target a default list title or ID on every tool call
- Designed for MCP-compatible AI clients that need hands-free backlog grooming during build and ship work
- Server schema version 0.4.0 on PyPI identifier google-tasks-mcp
- Stdio transport; optional SQLite DB_PATH for tokens and cache
- Three required secret/env vars: GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, GOOGLE_REDIRECT_URI
Community signal: 1 GitHub stars.
What problem does it solve?
Solo builders lose momentum when ship and launch todos sit in Google Tasks while the agent that could update them has no API access.
Who is it for?
Indie developers on Google Workspace who want agent-driven task capture during coding, review, and launch prep.
Skip if: Teams that need full project management, Kanban, or task sync without touching Google Cloud OAuth setup.
What do I get? / Deliverables
After install, your agent can create, list, and complete Tasks entries using your OAuth-connected account and optional default tasklist.
- Live Google Tasks read/write from agent tool calls
- Persistent OAuth token storage when DB_PATH is set
- Consistent default tasklist targeting when DEFAULT_TASKLIST is configured
Recommended MCP Servers
Journey fit
Canonical shelf is Build because the value is wiring Google Tasks into your agent stack as a live integration, not running a one-off research query. Integrations is the right subphase: OAuth-backed tasklist CRUD exposed as MCP tools for Claude Code, Cursor, and similar clients.
How it compares
MCP integration to Google Tasks, not a standalone task-management skill or Notion-style workspace.
Common Questions / FAQ
Who is Google Tasks MCP Server for?
It is for solo and indie builders using MCP agents who already rely on Google Tasks and want list operations from the same session where they write code.
When should I use Google Tasks MCP Server?
Use it when you are in the build or ship loop and need the agent to read or mutate your real task backlog instead of dumping reminders into chat.
How do I add Google Tasks MCP Server to my agent?
Install the PyPI package, register the stdio server in your client MCP config, and set GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, GOOGLE_REDIRECT_URI, plus optional DB_PATH and DEFAULT_TASKLIST.