Think inside your AI world.
How do I connect Snowflake to ChatGPT?
Snowflake's MCP path — Cortex Analyst for semantic querying, Cortex Search for retrieval — serves ChatGPT the governed warehouse in meaning.
Direct answer: connect Snowflake's MCP surface to ChatGPT. In ChatGPT, add it from the connector settings (developer mode); the tool's own docs carry the current path. Authorised, questions resolve through Cortex Analyst against governed data models, with Cortex Search for retrieval — semantic answers, not hand-rolled SQL.
Meaning-level queries, governed underneath
Revenue by segment, activation by cohort, the document that matches — asked in plain language, resolved against the models your data team governs. The governance travels with the answer.
Compute bills follow enthusiasm
Worked case: you ratified a warehouse-cost ceiling — the analytics warehouse suspends discussion-free scaling past a monthly credit line, after a quarter where exploratory enthusiasm doubled the bill and nobody could say which questions were worth it. Snowflake meters credits precisely; that a line exists where spending pauses for a conversation is financial law you set over the metering.
Through Unl the usage read arrives with the line drawn: credits at 86% of your ceiling mid-month — the conversation your rule schedules is due, doubled-bill story attached. Cost control becomes a standing read instead of an invoice ambush.
Snowflake to ChatGPT asks the governed warehouse in meaning; through Unl the usage answers to the credit ceilings you ratified — the cost conversation happens at 86%, not at the invoice.
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 is the Snowflake MCP surface?
Cortex Analyst (semantic queries over governed models) and Cortex Search (retrieval) — the same surface for every MCP client.
Does semantic querying change the governance story?
It strengthens it: questions resolve through governed models rather than raw tables, and your ratified rules in Unl sit above both — meaning governed twice, in different senses.
Why ratify a credit ceiling rather than set an alert?
Alerts notify; rulings frame. The ceiling arrives inside usage reads with its reasoning, so the pause-and-discuss happens as your policy, not as another ignored notification.
What if a single query takes you past the ceiling in one go?
Then the ceiling told you afterwards, which is the honest limit of a rule based on reading. It still beats the invoice by weeks. If you need it to stop the query rather than report it, that is a control inside Snowflake, and the ratified ceiling is what tells you where to set 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.