How do I connect Cursor to my warehouse?

Add Rig as an MCP server in Cursor. Rig builds the context layer over your warehouse and Cursor's agent can then query it with governed SQL.

Can Cursor write code against my real schema?

Yes. With Rig connected, Cursor reads your actual tables, joins and metric definitions and builds against them.

Is warehouse access read-only in Cursor?

Yes. Rig connects read-only and sandbox-validates each query before it runs.

Which warehouses work with Cursor through Rig?

Snowflake, BigQuery, Redshift, Postgres and DuckDB/MotherDuck, plus SaaS sources synced by Rig Ingest.

AI tools Supported

Cursor + Rig

The easiest way to query your warehouse data in Cursor

Cursor is an AI-native code editor that connects to external tools and data sources over MCP.

Try Rig Book a demo

What Rig gives Cursor

Once connected, Cursor can reason over your governed data through Rig:

Warehouse tables and views
Metric definitions
Join paths
dbt lineage
Role-based access
Sandbox-validated SQL

How Cursor connects to Rig

Add Rig as an MCP server in Cursor. The editor's agent can then query your warehouse with governed SQL, so the code it writes and the answers it gives are grounded in your real, validated data.

Write data-aware code in Cursor

With Rig connected, Cursor's agent reads your actual schema and metric definitions while it codes. It builds queries and components against real tables and correct joins instead of inventing column names.

Ask Rig

Build an API endpoint that returns weekly active users
Generate the SQL for revenue by cohort and wire it in
What columns does the orders table have?

Answer data questions without leaving the editor

Cursor can run ad-hoc data questions through Rig inline, returning validated numbers. You keep your flow in the editor rather than jumping to a separate analytics tool.

Ask Rig

How many signups did we get this week?
Show a sample of the subscriptions table
Find the join from accounts to invoices

Keep Cursor's queries governed and safe

Rig validates each query in a sandbox and runs read-only, so the editor's agent cannot touch production or run away with an unbounded scan. Answers use agreed metric definitions.

Ask Rig

Validate this query before running it
Recompute using the official active-user metric
Add a limit and re-run that against the warehouse

Prototype on real numbers, ship with confidence

Because the data Cursor sees is the governed warehouse, prototypes line up with production reporting. The metric in your prototype is the same metric the business uses.

Ask Rig

Mock this dashboard with live revenue numbers
Compare my calculation to the official metric
Explain how lifetime value is defined here

Frequently asked questions

How do I connect Cursor to my warehouse?: Add Rig as an MCP server in Cursor. Rig builds the context layer over your warehouse and Cursor's agent can then query it with governed SQL.
Can Cursor write code against my real schema?: Yes. With Rig connected, Cursor reads your actual tables, joins and metric definitions and builds against them.
Is warehouse access read-only in Cursor?: Yes. Rig connects read-only and sandbox-validates each query before it runs.
Which warehouses work with Cursor through Rig?: Snowflake, BigQuery, Redshift, Postgres and DuckDB/MotherDuck, plus SaaS sources synced by Rig Ingest.