microsoft-todo
MCP server from jordanburke/microsoft-todo-mcp-server
claude mcp add --transport stdio jordanburke-microsoft-todo-mcp-server npx --yes microsoft-todo-mcp-server \ --env MS_TODO_ACCESS_TOKEN="your_access_token" \ --env MS_TODO_REFRESH_TOKEN="your_refresh_token"
How to use
This MCP server exposes Microsoft To Do management via the Microsoft Graph API, enabling AI assistants to interact with your tasks, lists, and checklists. It provides a suite of tools (authentication, task lists, tasks, and checklist items) that support full CRUD operations and OData-style querying for tasks. After authenticating, the server stores tokens securely and automatically refreshes access tokens to keep the connection alive without manual intervention. You can generate an MCP configuration for your AI assistant (Claude, Cursor, etc.) so your assistant can create, read, update, and delete to-dos as part of its conversation flows. To use it, configure your AI assistant to point to the MCP server endpoint via an mcp.json entry, then ensure the environment contains valid tokens or a valid OAuth flow has completed. The server supports multi-tenant scenarios and uses TypeScript with ES modules for a modern, typed development experience.
How to install
Prerequisites:
- Node.js 16 or higher (test with Node.js 18.x, 20.x, 22.x)
- pnpm package manager
- A Microsoft account (personal, work, or school)
- Azure App Registration with Redirect URI configured
Option 1: Global Installation (Recommended)
# Install globally using npm
npm install -g microsoft-todo-mcp-server
# Or using pnpm
pnpm install -g microsoft-todo-mcp-server
# Or run directly with npx (no installation)
npx microsoft-todo-mcp-server
The package provides three command aliases:
- microsoft-todo-mcp-server
- mstodo
- mstodo-config
Option 2: Clone and Run Locally
git clone https://github.com/jordanburke/microsoft-todo-mcp-server.git
cd microsoft-todo-mcp-server
pnpm install
pnpm run build
Azure App Registration steps are described in the repository README to obtain client credentials and configure permissions. After registration, set up environment variables (CLIENT_ID, CLIENT_SECRET, TENANT_ID, REDIRECT_URI) and prepare a tokens.json via the authentication flow when running the server.
Additional notes
Tips and common considerations:
- Tokens.json is stored with automatic refresh; you can override the token file location using MSTODO_TOKEN_FILE or pass tokens via environment variables (MS_TODO_ACCESS_TOKEN, MS_TODO_REFRESH_TOKEN).
- Use the provided helpers (mstodo-config or create-config) to generate the mcp.json configuration for your AI assistant.
- The MCP server exposes 15 tools for managing lists, tasks, and checklist items, along with authentication status checks. Ensure your OAuth setup includes the necessary Microsoft Graph permissions (Tasks.Read, Tasks.ReadWrite, User.Read).
- When running with npx or a global install, ensure network access to fetch the package and dependencies. For local development, use pnpm install and pnpm run dev or pnpm run start as appropriate.
- If integrating with Cursor or Claude Desktop, export or mount the mcp.json in the AI assistant’s expected config path as shown in the README, and provide the access/refresh tokens in the environment or tokens.json generated by the auth flow.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud