Think inside your AI world.

MotherDuck through Unl

MotherDuck's MCP server will list every column a data share carries, right down to the field names. Whether one of those fields is the one your contract says can never leave the building is a call your own rule makes - a share that runs isn't the same as a share that's cleared.

MotherDuck through Unl reads a share's table and column list against the partner-contract rule you have ratified, so 'is this share ready to send?' comes back qualified with the condition it hasn't met.

What MotherDuck holds

Ask MotherDuck's MCP server about a database or share and it returns:

  • database, table, and column listings, including types and comments
  • fuzzy search across databases, schemas, tables, columns, and shares
  • the list of database shares sent to or from your account
  • SQL query results run directly against MotherDuck databases
  • answers to documentation questions via the built-in docs tool

What the naked read gives you

MotherDuck's MCP server lists every database, table, and column on request, fuzzy-searches the whole catalogue, and lists the shares sent to or from an account - down to which column carries which type and comment. It runs the SQL and hands back the result. What it does not do is say whether a column sitting inside a share is one that share is supposed to carry.

The frame judges the data it is given; it does not verify the source’s accuracy.

What changes when MotherDuck is read measured

A data platform lead at an analytics vendor ratified a share rule in Unl: no data share goes out to a partner while its column list still includes a field tagged customer_email, because a share sent eight months ago carried a raw contact column into a partner's warehouse and triggered a client audit.

'Is this share ready to send to the partner?' returns, measured: not yet - the share's column read shows customer_email still present in the orders_summary table, the exact field the contract excludes. MotherDuck supplied the column list; your ratified rule supplied the exclusion.

And back again

Drop the flagged column from the shared view, and the next read shows a clean list; MotherDuck through Unl updates this same page once the ratified exclusion rule sees it.

The answer comes back measured against what you already decided, and why.

The lane is live and open to this tool today: 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.

Questions people ask

What would clear this share to go out?

Removing customer_email from the shared view, or replacing it with a masked column. The same ratified rule reads the updated column list and clears the share.

How do I connect MotherDuck to Claude?

Add MotherDuck's managed remote MCP server (api.motherduck.com/mcp) in Claude's MCP settings, authenticating with a MotherDuck access token, or run the local mcp-server-motherduck package against the same token - either way Unl reads through that connection.

Does Unl write to my database?

Unl reads through MotherDuck, and can write back on your explicit gesture — it never acts as a side effect of a read.

What this is

Think inside your AI world — you stay in command

Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.

It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.

Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:

Alpha is invite-only · free at launch.