Snowflake metadata integration for Rig
The easiest way to query Snowflake metadata in Claude and other AI tools
Snowflake metadata is a read-only view of a Snowflake account's structure: its tables, columns, view definitions and recent SQL history.
What Rig sees in Snowflake metadata
Once connected, Rig models these objects so questions resolve to the right tables and joins:
- Tables
- Columns
- View definitions
- Query history
How Snowflake metadata connects to Rig
Rig signs in to Snowflake with a key pair and reads the schemas you allow. No business rows are copied, so Rig learns the structure and how people query it while the data stays where it is.
Read the step-by-step Snowflake metadata setup guideMap what is in your Snowflake warehouse
A catalogue is the list of tables and columns in a warehouse. Rig reads yours for the schemas you allow, so you can ask what exists, how columns are typed and which tables belong to which area without opening Snowflake.
- Which tables in the ANALYTICS schema have a customer_id column?
- List the columns and types of the orders table
- How many tables does each allowed schema hold?
Understand how views are built
A view definition is the SQL behind a derived table. Rig reads the definitions for your allowed schemas, so you can trace where a reporting table comes from and which tables it depends on.
- Show the SQL behind the monthly_revenue view
- Which views read from the raw_orders table?
- Which views in MARTS join more than three tables?
Learn from the SQL your team already runs
Query history is the record of SQL run in your account. Rig reads a bounded window of it, 30 days by default, so it can see which tables and joins people actually use and answer questions the same way.
- Which tables were queried most in the last 30 days?
- What joins do analysts use between orders and customers?
- Which tables have not appeared in any query recently?
Ask your Snowflake metadata data in Claude, ChatGPT or Cursor
Rig serves your Snowflake metadata data to AI assistants over MCP. Once Snowflake metadata is synced into your warehouse, you can ask questions in plain English from the tool you already use, and every answer is backed by validated, governed SQL.
Frequently asked questions
- Does Rig copy my Snowflake data?
- No. The connection reads structure and SQL text only. No business rows are copied.
- How does Rig sign in to Snowflake?
- With key-pair authentication. You assign a public key to a Snowflake user, and Rig holds the private key.
- Can I limit which schemas Rig sees?
- Yes. You list the allowed schemas when you connect, and Rig stays inside that list.
- How much query history does Rig read?
- The last 30 days and up to 100,000 rows by default. Both limits are settings on the connection.
Related integrations
Connect Snowflake metadata data to AI tools like Claude
Rig builds a governed context layer over your data so every team, and every AI tool, asks questions and gets answers they can trust.