Think inside your AI world.
BigQuery through Unl
BigQuery will return exactly what you query. Whether that number is the number depends on a definition you settled — and a query can quietly compute a different one.
BigQuery through Unl reads your live warehouse — query results, metadata, resources — against the metric definition you have ratified, so the answer is measured against the meaning you settled, with any mismatch to the query’s own logic flagged.
What BigQuery holds
BigQuery’s official remote MCP server exposes the warehouse surface:
- Query execution and results
- Table and dataset metadata
- Resources
What the naked read gives you
A naked read runs the query and returns the number. Accurate to the SQL, and silent on whether that SQL matches your ratified definition — because the definition lives with you, not in the query.
The frame judges the data it is given; it does not verify the source’s accuracy.
What changes when BigQuery is read measured
Say you’ve ratified the north-star definition: an active account is one with a qualifying event in the trailing 28 days, deduped by organisation — the meaning the board signed off on.
“How many active accounts?” returns, measured: on your ratified definition, 1,240 — and a flag: the dashboard query uses a 30-day window and counts users, not organisations, so its 1,610 isn’t your metric. The gap is the definition, and yours is the one that was agreed.
And back again
When you update the window to 30 days across the board, that revised definition is settled once in Unl — and the next query read is measured against the meaning the team now holds.
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.
Read further
Questions people ask
Can AI make a BigQuery result match our agreed metric?
Yes, through Unl. You ratify the metric definition, and the BigQuery result is measured against it — the answer holds the agreed meaning and flags where a query’s own logic diverges from your definition.
How do I connect BigQuery to Claude?
BigQuery offers an official remote MCP server for queries and metadata. Through Unl the same results arrive measured against your ratified definition, so a number reads as your metric, not just SQL output.
Does Unl write to my warehouse?
Unl reads through BigQuery, and can write back on your explicit gesture — it never acts as a side effect of a read.
What if two subsidiaries of one customer are separate organisations in the warehouse?
Then they count twice, because the dedupe runs on the organisation field the warehouse holds. Which legal entities are one customer is a definition you carry; BigQuery applies the key it has. The read tells you the key it deduped on, so a north-star number is never presented as more settled than the field behind 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.