Doist/todoist-mcp

What it solves

This project provides a bridge between AI agents and Todoist, allowing LLMs to access and modify a user's tasks and account information. It eliminates the need for manual data entry by enabling AI assistants to manage productivity workflows directly within a task management system.

How it works

It implements the Model Context Protocol (MCP), which provides a standardized way for AI clients (like Claude Desktop, Cursor, or VS Code) to connect to external tools. The project can be deployed as a streamable HTTP service or imported as a library into custom AI applications using SDKs like Vercel's AI SDK. It includes a set of tools that wrap the Todoist API, allowing the AI to perform actions such as finding tasks by date or adding new tasks.

Who it’s for

• AI Agent Developers: Developers building conversational interfaces that need task management capabilities.
• AI Power Users: Users of MCP-compatible clients (Claude, Cursor, VS Code) who want their AI assistant to manage their Todoist account.

Highlights

• Multi-Client Support: Compatible with Claude Desktop, Cursor, Claude Code (CLI), and Visual Studio Code.
• Flexible Integration: Can be used as a standalone MCP server or as a direct library dependency via npm.
• OpenAI Compatibility: Includes tools optimized for the OpenAI MCP specification.
• Interactive UI: Supports "MCP Apps," which are interactive UI widgets that render tool outputs visually (e.g., as task lists) instead of plain text.