You can connect Todoist to Claude in a few minutes using the Model Context Protocol (MCP), and once it’s connected, Claude can create, search, and complete tasks and projects without you switching over to the app to do it by hand. Todoist’s OAuth also just works with a new client, the same as Notion, Airtable, and Neon in this series. This guide covers both ways to do it: Todoist’s official remote MCP server connected directly, and connecting it through a gateway if you use more than one AI client.

What the Todoist MCP server actually does

Todoist publishes an official remote MCP server at ai.todoist.net/mcp. It exposes your tasks and projects as tools Claude can call directly, so prompts like these work against your real Todoist account:

  • “Add a task to follow up with the vendor on Friday.”
  • “What’s still open on my Q3 launch project?”
  • “Mark yesterday’s three overdue tasks as done.”
  • “Turn these five bullet points from our call notes into tasks.”

Claude acts through the same permissions your Todoist account already has — it isn’t a read-only summary or an export you review afterward, it’s your account creating, editing, and completing tasks in place, the same as if you’d typed them in yourself.

Option 1: Connect Todoist to Claude directly

Claude supports remote MCP servers natively. The short version of the setup:

  • Open Claude’s connector or MCP settings.
  • Add a remote server pointing at Todoist’s MCP endpoint.
  • Sign in with your Todoist account when prompted — this is standard OAuth, so Claude never sees your Todoist password, and you can revoke access from Todoist’s own account settings at any time.

That’s enough if Claude is the only AI client you use with Todoist. The tradeoffs show up once you add a second tool or a second teammate: every additional AI client (ChatGPT, Cursor, an agent) needs its own copy of that same connection, its own OAuth grant, and its own place where the config can go stale.

Option 2: Connect it once, through a gateway

If you’re already juggling more than one MCP server or more than one AI client, wiring each pair directly gets old fast — that’s the connection sprawl we wrote about in what an MCP gateway is. The alternative is connecting Todoist (and everything else) to one gateway URL, then pointing every AI client at that single URL instead of reconfiguring each one.

With gate specifically, connecting Todoist looks like this: add gate’s gateway URL to Claude once, then pick Todoist from the Todoist server page in the directory and sign in with OAuth the same way you would directly. From then on, Todoist’s tools are available in every client connected to your gate URL — Claude, ChatGPT, or Cursor — without a second setup step. See the full list of servers set up this way on the MCP servers page.

Why the connection method still matters for a task manager

A task list is lower stakes than a payments account or a production database, but “lower stakes” isn’t “no stakes.” A few things worth checking before or right after you connect:

  • Is it the official endpoint? Connect Todoist’s documented server (ai.todoist.net), not a third-party mirror claiming to proxy it.
  • What can it actually write? A tool that can create and complete tasks can also delete a project or clear a list you didn’t mean to touch — worth knowing before you hand it an open-ended prompt like “clean up my backlog.”
  • Shared projects change the blast radius. If your Todoist account has projects shared with teammates or clients, a task Claude creates or edits there is visible to everyone on it, not just you.
  • Who else on the team can reach it? If more than one person connects Todoist to their own AI client, a shared, revocable setup beats everyone holding their own grant — the same access-control question covered in MCP access control for teams.

You can run a quick, no-signup check on any server’s tool list with gate’s free MCP security scanner, which looks for prompt injection and hidden instructions in tool descriptions before you rely on a server day to day.

Troubleshooting the connection

A few things that commonly go wrong when connecting any remote MCP server, Todoist included:

  • OAuth loop or sign-in fails. Make sure you’re signed into the right Todoist account — and the right workspace, if you use Todoist for both personal and team projects — before authorizing.
  • Claude can’t find a task or project you know exists. Todoist’s tools respect your account’s own permissions; a project shared with you under a different account won’t show up while Claude is connected to another one.
  • Tools disappear after a Todoist update. Remote servers can change their tool list without warning. If you’re connected through a gateway, that kind of change is exactly what drift detection is meant to catch — see how MCP rug pulls work for why a one-time review isn’t enough.
Where gate fits: gate connects Todoist once, behind a URL every AI client can share, with allow/ask/block rules you can set per tool — so listing tasks can run automatically while deleting a project waits for a human. Every server is scanned for prompt injection before it goes live, and every call is recorded in a plain-language log. See the full gateway at /mcp-gateway.

The bottom line

Connecting Todoist to Claude with MCP takes a few minutes either way: point Claude directly at Todoist’s server and sign in, or connect it once through a gateway if you want the same access available in every AI client you use, with rules on what it can do without asking first. It’s one of the lower-stakes connections in this series, but a tool that can silently clear a shared project is still worth a minute’s thought before you say yes.