Think inside your AI world.
How do I connect Asana to Claude?
Asana exposes its Work Graph — tasks, projects, goals — over MCP, so Claude can query the plan instead of you paraphrasing it.
Direct answer: add Asana's MCP server in Claude via Settings → Connectors and authorise; Claude Code registers it with 'claude mcp add'. The read surface is the Work Graph: tasks, projects and goals, connected the way Asana connects them, so questions about who owns what, where a project stands and how work rolls up to goals are answered from the live graph.
Asking the plan questions
Which tasks block the launch project, what is due this week across clients, how the quarter's goal is tracking against its projects — read straight from the graph. The goals layer is Asana's quiet strength here: work arrives already attached to intent.
The ceiling the graph never sees
A studio example: you ratified a work-in-progress limit — no more than three client projects live at once — after a five-project month taught everyone the cost of context thrash. Asana displays all five columns without complaint. Limits are choices about the graph; the graph has no slot for choices.
The graph, measured
Through Unl the project read lands against your limit: four live — one over the ceiling, and the ceiling has its month of scar tissue attached. Capacity conversations start from a verdict instead of a recount. The rule earns its keep every time someone proposes 'just one more'.
Asana in Claude tells you what work exists; through Unl it tells you whether the shape of that work honours the limits you set for yourselves, with the history that set them.
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 the Asana MCP server give Claude?
The Work Graph read: tasks, projects and goals with their connections — enough to answer ownership, status and roll-up questions conversationally from live data.
Does this work outside Claude?
Yes — the server is standard MCP, addable by ChatGPT and other MCP clients. This page's steps translate directly; only the client's connector menu differs.
What is worth ratifying for a services team?
WIP ceilings, client-response norms, a definition of project-done that includes handover. Each is a limit with a lesson behind it — exactly the unit Unl holds and serves when the graph read touches it.
What if the limit we set was for a client we no longer have?
Supersede it. A limit nobody revisits is how a rule written for one engagement starts blocking work it was never about — and because the original carries its reasoning, you can see at a glance whether the reason still applies before you retire it.
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.