Advertise
MCP.so
Sign In

Todoist MCP Server (Python)

@mikemc

Repository

Basic information

Category

Productivity

License

MIT

Transports

stdio

Author

mikemc

About Todoist MCP Server

Todoist MCP server

Todoist MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "todoist": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/mikemc/todoist-mcp-server",
        "todoist-mcp"
      ],
      "env": {
        "TODOIST_API_TOKEN": "your_todoist_api_token"
      }
    }
  }
}

Todoist MCP Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

Fetch the live tool list by running this server in a temporary sandbox using the button above.

Overview

What is Todoist MCP Server (Python)?

A Todoist MCP server written in Python that uses the Todoist Python API to allow AI assistants (like Claude and Goose) to manage Todoist tasks, projects, sections, and comments. It was initially created by translating a TypeScript Todoist MCP server to Python.

How to use Todoist MCP Server (Python)?

Install Python 3.10+, the UV package manager, and obtain a Todoist API token. Configure the server in claude_desktop_config.json using the uvx command with your token. Alternatively, configure it with Goose via goose configure and setting environment variables. Then invoke tools like todoist_get_tasks, todoist_add_task, etc.

Key features of Todoist MCP Server (Python)

  • Full CRUD for projects, sections, tasks, and comments
  • Filter and search tasks by criteria
  • Complete and uncomplete tasks
  • Move tasks between sections or projects
  • Integrates with Claude Desktop and Goose (local LLM)

Use cases of Todoist MCP Server (Python)

  • Query completed tasks from the last 7 days
  • Add new tasks to Todoist via AI chat
  • Update project or section names through natural language
  • Retrieve and filter tasks by label, priority, or date
  • Delete or move tasks without opening the Todoist app

Frequently asked questions

1

What tools are available?

The server provides tools for projects, sections, tasks, and comments, including `todoist_get_projects`, `todoist_add_task`, `todoist_complete_task`, `todoist_get_comments`, and more.

2

How do I get a Todoist API token?

Log in to your Todoist account, go to Settings → Integrations, and find your API token under "Developer".

3

Can I use this server with a local LLM?

Yes, you can configure it with Goose using a local model provider like LM Studio or Ollama.

4

What are the runtime dependencies?

The server requires Python 3.10+, the UV package manager, and a Todoist API token. No separate database or runtime is needed beyond these.

5

Where does the code come from?

This project is a Python translation of the TypeScript [todoist-mcp-server](https://github.com/abhiz123/todoist-mcp-server) by abhiz123, and is licensed under the MIT License.

Comments

More Productivity MCP servers

adminlanding

Julien Maurice

AdminLanding — French rental, frontalier & home-employment compliance tools (FR/EN). Sourced, dated answers from deterministic rule engines and official datasets — no AI, no account needed for the 13 free tools: zone ten

WhatsMCP

whatsmcp

WhatsMCP — WhatsApp for AI agents. Put an AI agent on a real WhatsApp number. Expose regular WhatsApp accounts to your LLM agents over JSON-RPC. A complete Model Context Protocol server that turns phone numbers into addr

Genviral

Genviral

Plan, create, schedule, publish, and analyze social media content from AI assistants through Genviral's secure OAuth MCP server.

Dadan MCP

Dadan-io

dadan-sdk-integration is an SDK for integrating Dadan features into React applications. It allows you to add video recording and file selection functionality directly into your React components, providing a seamless user

twocents

twocents-page

Human feedback for AI agents — remote MCP server at twocents.page/mcp (share_page / get_feedback)

Advertise