Think inside your AI world.
How do I connect Todoist to ChatGPT?
A personal list over MCP answers from wherever you think — and ChatGPT is where a lot of thinking happens.
Direct answer: connect Todoist over MCP to ChatGPT. In ChatGPT, add it from the connector settings (developer mode); the tool's own docs carry the current path. Authorised, ChatGPT reads your projects, tasks and due dates live — the week ahead, the overdue tail, the project that stopped moving — from the account, not from your recollection of it.
Reviews as conversations
What does Friday actually require, what has rolled over three times, which project gained ten tasks and finished none — the weekly review, conducted by asking. The overdue tail never lies, but it does hide; retrieval un-hides it.
The someday list needs a fence
Worked case: you ratified a someday-cap — the someday/maybe list holds thirty items, pruned monthly, because an unbounded list of 212 items once became a guilt archive nobody opened. Todoist counts items; that thirty is where aspiration stays honest is your fence around your own attention.
Through Unl the review read arrives fenced: someday at thirty-eight — eight over your cap, prune due, and the 212-item archive is the reason the fence exists. The list stays a list instead of becoming sediment.
Todoist to ChatGPT reads the list; through Unl it reads the list against the fences you ratified — the someday cap holds because it shows up at review time with its story.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
What does ChatGPT read from Todoist?
Projects, tasks and due dates from your account — the same live surface as any MCP client connection.
Is a personal task list worth the measured layer?
Personal systems fail on judgement, not data: caps, fences, honest definitions of 'active'. Those are exactly the units Unl holds — small rules with reasons, applied at every review.
Can I use both Claude and ChatGPT against the same list?
Yes — one server, many clients, and the rules in Unl are client-independent. The cap is the cap wherever you review from.
What if you keep raising the cap at every review?
Then the cap is not a fence, it is a formality, and the record shows that plainly because each raise is its own decision with a date. That trail is more useful than the number. A cap you have lifted four times is telling you something about the list that the list itself will not say.
What this is
Think inside your AI world — you stay in command
Save the thoughts, decisions and targets worth keeping, each with its reasoning, carried into every AI session the moment they matter. A new unit of exchange between you and your AI: the Settled Why with standing that travels. Unprompted.
Your whole AI world. What you decided at the epicentre. It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector — quick to connect, in a couple of steps.
MCP native·Human settled·Model agnostic·Your data
Measured Context
Connect Unl to bring the right information into the moment.
Your sources, read against the criteria you set.
The full product, open. Free at launch.