AnotherTodo is a todo list. It’s also an MCP (Model Context Protocol) server. Here’s how to connect it to Claude and start managing your todos from natural language.

What you get

Five tools. That’s it.

• list_todos — see all your todos
• add_todo — add a new one
• update_todo — change the text
• delete_todo — remove it
• toggle_todo — mark done/undone

Claude calls these tools when you ask it to do things like “add buy milk to my todo list” or “what’s on my list?”

Setup: Claude Desktop

1. Sign up at anothertodo.app and subscribe ($4/mo)
2. Copy your MCP URL from the dashboard — it looks like https://anothertodo.app/mcp/tmcp_your_key_here
3. Open Claude Desktop settings → MCP Servers → Add
4. Paste your URL as a Streamable HTTP server
5. Done. Ask Claude to list your todos.

Setup: claude.ai

1. Go to Settings → Integrations → Add MCP Server
2. Paste your MCP URL
3. That’s it. Claude now has access to your todo list.

Setup: Claude API

If you’re building something with the Claude API, you can connect AnotherTodo as an MCP tool source. Pass the MCP URL in your tool configuration.

What happens next

You say “add take out the trash” and Claude calls add_todo with the text “take out the trash.” It appears in your web dashboard and in every future list_todos call.

Is this the most efficient way to manage a todo list? No. Is it the most fun? Also no. But it works, and there’s something satisfying about telling an AI to manage your groceries.

Rate limits

60 requests per minute per API key. If you hit that managing a todo list, we need to talk about your productivity system.