How do I connect Claude Code to my warehouse?

Add Rig as an MCP server in Claude Code. Rig builds the context layer over your warehouse, and Claude Code can then query it with governed SQL.

Can Claude Code build dashboards on live data?

Yes. With Rig connected, Claude Code writes code against your real schema and metric definitions and pulls validated numbers as it builds.

Is the data access read-only?

Yes. Rig connects read-only and sandbox-validates each query, so generated SQL cannot mutate production data.

Which warehouses are supported?

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

AI tools Supported

Claude Code + Rig

The easiest way to query your warehouse data in Claude Code

Claude Code is Anthropic's agentic coding tool that runs in the terminal and IDE and connects to external tools and data over MCP.

Try Rig Book a demo

What Rig gives Claude Code

Once connected, Claude Code 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 Claude Code connects to Rig

Add Rig as an MCP server in Claude Code. It can then query your warehouse with governed SQL while you build, so dashboards, scripts and analyses are grounded in real, validated data rather than made-up columns.

Vibe-code dashboards on real warehouse data

With Rig connected, Claude Code writes app and dashboard code against your actual schema and metric definitions. It pulls correct joins and real numbers as it builds, so the chart you get reflects the warehouse rather than placeholder data.

Ask Rig

Build a retention dashboard from our warehouse data
Add a revenue-by-region chart using the official metric
Wire this component to the live orders table

Investigate data straight from the terminal

Claude Code can answer ad-hoc data questions inline while you work, generating validated SQL through Rig. You stay in the terminal instead of switching to a SQL client to check a number.

Ask Rig

How many active accounts do we have right now?
Show the schema and a sample of the events table
Find the join path from users to subscriptions

Keep generated SQL safe and correct

Every query Claude Code runs through Rig is sandbox-validated and read-only, so an exploratory question never mutates production or runs an unbounded scan. Answers use your agreed metric definitions.

Ask Rig

Validate and run this query against the warehouse
Recompute that with the official churn definition
Limit the result and re-run it

Turn analysis into a committed workflow

Because Rig governs the data path, analyses you build in Claude Code can be promoted into repeatable workflows. The same context that answers a question can power a scheduled report or a downstream action.

Ask Rig

Turn this analysis into a scheduled weekly report
Write a script that flags at-risk accounts nightly
Document the metrics this dashboard uses

Frequently asked questions

How do I connect Claude Code to my warehouse?: Add Rig as an MCP server in Claude Code. Rig builds the context layer over your warehouse, and Claude Code can then query it with governed SQL.
Can Claude Code build dashboards on live data?: Yes. With Rig connected, Claude Code writes code against your real schema and metric definitions and pulls validated numbers as it builds.
Is the data access read-only?: Yes. Rig connects read-only and sandbox-validates each query, so generated SQL cannot mutate production data.
Which warehouses are supported?: Snowflake, BigQuery, Redshift, Postgres and DuckDB/MotherDuck, plus SaaS sources synced through Rig Ingest.