Todoist MCP Server Extended
An MCP (Model Context Protocol) server implementation that integrates Claude - or any MCP compatible LLM if you’re crafty - with Todoist, enabling natural language task management via MCP tools. The tools in this server allows Claude to interact with your Todoist tasks, projects, sections, and labels using everyday language, while also optimized for LLM workflow efficiency.
Features Overview
- Task Management: Create, update, complete, and delete tasks using everyday language
- Label Management: Create, update, and manage personal labels and task labels
- Project Management: Create, update, and manage Todoist projects
- Section Organization: Create and manage sections within projects
- Smart Search: Find tasks and labels using partial name matches
- Flexible Filtering: Filter tasks by project, section, due date, priority, and labels
- Rich Task Details: Support for descriptions, due dates, priority levels, and project/section assignment
- Batch Operations: Tools have built in batch operation support and custom parameters for efficient usage with LLM workflows
For a complete list of available tools as well as their usage, see tools.md.
Quick Installation Guide
Assuming you already have npm installed.
A more comprehensive installation guide can be found in the How-to Guide.
Installing via Smithery
To install Todoist MCP Server Extended for Claude Desktop via Smithery:
- Run following command in cmdpwsh:
npx -y @smithery/cli install @Chrusic/todoist-mcp-server-extended --client claude
Also compatible with cline or windsurf, by changing last parameter to --client cline
or --client windsurf
Installing via npm
- Run following command in cmdpwsh:
npm install -g @chrusic/todoist-mcp-server-extended
Setup
Grab a Todoist API Token
- Log in to your Todoist account
- Navigate to
Settings → Integrations
- Find your API token under
Developer
- Press
Copy API Token
For more information about the Todoist API, visit the official Todoist API documentation.
Add MCP Server and API Token Claude Desktop Client
In your
claude_desktop_config.json
file, paste the following json snippet between:"mcpServers":{ }:
"todoist": { "command": "npx", "args": ["-y", "@chrusic/todoist-mcp-server-extended"], "env": { "TODOIST_API_TOKEN": "PASTE-YOUR-API-TOKEN-HERE" } }
When all put together, it should look something like this:
{ "mcpServers": { "todoist": { "command": "npx", "args": ["-y", "@chrusic/todoist-mcp-server-extended"], "env": { "TODOIST_API_TOKEN": "PASTE-YOUR-API-TOKEN-HERE" } } } }
Claude Desktop client will then start the MCP server and load the tools on the next client (re)start.
Example Usage
Some simple suggestions on what to ask Claude. Note that sometimes you have to be very direct to get claude to use the tools:
- “Using the MCP tool: todoist_get_tasks, list all my tasks for the day.”
- “Create task ‘Review PR’ in project ‘Work’ section ‘To Do’”
- “Add label ‘Important’ to task ‘Review PR’”
- “Show all tasks with label ‘Important’ in project ‘Work’”
- “Move task ‘Documentation’ to section ‘In Progress’”
- “Mark the documentation task as complete”
- “Give me some suggestions for listed tasks I can do today as I’m going shopping in town.”
- “Break task X down in to smaller subtasks and add due dates, x, y, z.”
Contributing
Contributions are welcome! Feel free to submit a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Issues and Support
If you encounter any issues or need support, please file an issue on the GitHub repository.
Todoist-mcp-server-extended
Project Details
- Chrusic/todoist-mcp-server-extended
- @chrusic/todoist-mcp-server-extended
- MIT License
- Last Updated: 4/11/2025
Recomended MCP Servers
Airtable integration for AI-powered applications via Anthropic's Model Context Protocol (MCP). Connect your AI tools directly to Airtable...
The core MCP extension for Systemprompt MCP multimodal client
cocosMCP
Bring your project into LLM context - tool and MCP server
MCP server for Delve debugger integration
AI-powered search capabilities for AI assistants using the Tavily API and Model Context Protocol (MCP)
Run any AWS Lambda function as a Large Language Model (LLM) tool without code changes using Anthropic's Model...
This is an MCP server that interacts with a PocketBase instance. It allows you to fetch, list, create,...
The Okta MCP Server is a groundbreaking tool built by the team at Fctr that enables AI models...
Tool to work with arXiv, provide LLM with ability to search and read papers from there