todo
A simple, powerful Todo list manager for Claude Desktop and other MCP-compatible AI assistants. Organize your tasks across different projects with priorities and never lose track of what needs to be done!
claude mcp add --transport stdio floriscornel-todo-mcp npx -y @floriscornel/todo-mcp@latest \ --env DATABASE_URL="postgresql://postgres:postgres@localhost:5432/todo_mcp"
How to use
Todo MCP provides a persistent task management assistant that integrates with Claude Desktop and other MCP-compatible AI assistants. It organizes tasks across multiple lists with priorities, timestamps, and persistent storage in PostgreSQL. You can interact with Claude to create lists, add tasks, assign priorities (low, medium, high, urgent), and query or complete tasks. The included example config shows how to wire the MCP into Claude Desktop, using npx to fetch the latest package and connecting to a PostgreSQL database via the DATABASE_URL environment variable. Use natural language prompts to manage tasks, e.g., creating lists like 'Frontend Project', adding tasks with priorities, and filtering by urgency or due dates.
How to install
Prerequisites:
- Node.js (16.x or newer) and npm installed on your system
- PostgreSQL server running and accessible
- Basic familiarity with command-line tools
Step 1: Install and start PostgreSQL
- If you’re running locally, install PostgreSQL and create the needed databases (todo_mcp, todo_mcp_test).
Step 2: Install the MCP server via npx
- The MCP server is published as @floriscornel/todo-mcp. You can run it directly via npx:
# Install and run the latest Todo MCP package
npx -y @floriscornel/todo-mcp@latest
Step 3: Configure the database connection
- Ensure PostgreSQL is accessible and set the DATABASE_URL environment variable to point to your database:
export DATABASE_URL=postgresql://postgres:postgres@localhost:5432/todo_mcp
Step 4: Integrate with Claude Desktop (example)
- In Claude Desktop MCP settings, configure:
{
"mcpServers": {
"todo-mcp": {
"command": "npx",
"args": ["-y", "@floriscornel/todo-mcp@latest"],
"env": {
"DATABASE_URL": "postgresql://postgres:postgres@localhost:5432/todo_mcp"
}
}
}
}
Step 5: Run and verify
- Start Claude Desktop and select the Todo MCP server configuration. Test creating lists, adding tasks with priorities, and querying urgent tasks.
Additional notes
Notes and tips:
- The PostgreSQL database stores all tasks, lists, and metadata; ensure the database URL is correct and that the user has appropriate permissions.
- You can override the default DATABASE_URL with a different host/port/database as needed for different environments (development, testing, production).
- For testing, you may create a separate database like todo_mcp_test as shown in the README.
- If you encounter connection issues, verify that the PostgreSQL server is running and reachable from your host, and that the credentials match your setup.
- This MCP exposes a range of task-management capabilities via natural language prompts; use phrases like: "Create a list called 'Frontend Project'", "Add a high-priority task to fix login bug", "Show me all urgent tasks".
Related MCP Servers
obsidian -tools
Add Obsidian integrations like semantic search and custom Templater prompts to Claude or any MCP client.
mcp-streamable-http
Example implementation of MCP Streamable HTTP client/server in Python and TypeScript.
Matryoshka
MCP server for token-efficient large document analysis via the use of REPL state
Agentic -Skill
Agentic-MCP, Progressive MCP client with three-layer lazy loading. Validates AgentSkills.io pattern for efficient token usage. Use MCP without pre-install & wasting full-loading
pinmeto-location
PinMeTo MCP server that enables users with authorized credentials to unlock their data
mongo
MCP server that provide tools to LLMs such as claude in cursor to interact with MongoDB