Skip to content

Start building

Integration · Databases

AI agents for Snowflake

Run read-only SQL in Snowflake and page through the results, and list databases, schemas, tables, columns and warehouses, through the SQL API with a programmatic access token or a key pair.

Actions
9
Read
8
Write
1

Actions

What your agents can do in Snowflake

9 actions. Each one is a tool your agents can call and a step your workflows can run.

Read 8

Look things up. Nothing changes.

  • Run a SQL query

    Runs one read-only SQL statement (SELECT, WITH, SHOW, DESCRIBE, DESC or EXPLAIN; only one statement) and waits up to timeout_s for it.

  • Get query results

    Checks a statement by its statementHandle and reads its results: the first partition, or partition N for a large result (count them in resultSetMetaData.partitionInfo).

  • List databases

    Lists the databases the role can see (SHOW DATABASES).

  • List schemas

    Lists the schemas of a database (from its INFORMATION_SCHEMA).

  • List tables

    Lists the tables and views of a database, or of one schema (names match without regard to case).

  • Describe a table

    Lists a table's or view's columns (names match without regard to case).

  • Preview a table

    Reads the first rows (10 by default, at most 1,000) of a table or view by its name, like SALES.PUBLIC.ORDERS or ORDERS with a default database and schema.

  • List warehouses

    Lists the warehouses the role can see (SHOW WAREHOUSES).

Write 1

Create, change or send.

  • Cancel a running query

    Cancels a statement that is still running, by its statementHandle.

A person can approve first

Any write action can be set to wait for a person's approval before it runs, so an agent asks before it changes anything.

Connection

How it connects

Connect it once in the console, then give it to the agents and workflows that need it.

Sign-in details

Enter the sign-in details Snowflake gives you for its API.

What you enter

  • Programmatic access token or private key

Settings

  • Account identifier

Kept safe

  • Credentials are stored encrypted with your organization's own keys.
  • They are opened only by the process that makes the call, at the moment it runs, and never go into a prompt or the browser.
  • Calls go only to the tool's own servers.

How we protect your data

Ideas

Ideas to start with

Starting points made only from the actions above. Mix them with any other tool your agents can use.

  • 01

    Ask about your Snowflake data

    Ask a question in plain words. The agent can run a SQL query, get query results or list databases in Snowflake, then answers with what it found.

  • 02

    Let an agent update Snowflake

    Have the agent cancel a running query in Snowflake as part of a larger task. Turn on approval and a person signs off on each change first.

  • 03

    Share what it finds in Slack

    Give one agent Snowflake and Slack: it can run a SQL query in Snowflake and post a summary to a Slack channel in the same run.

FAQ

Snowflake and MyDeepAgents

Can MyDeepAgents connect to Snowflake?

Yes. The Snowflake connector has 9 actions: 8 read and 1 makes changes. Your agents can call them as tools, and your workflows can run them as steps.

How do I connect Snowflake?

Enter the sign-in details Snowflake gives you for its API. You'll need: Programmatic access token or private key. The connection also asks for: Account identifier. Credentials are stored encrypted with your organization's own keys and are opened only at the moment of a call.

Can an agent change data in Snowflake without asking?

Only if you let it. Its one action that changes data can be set to wait for a person's approval before it runs.

Can I use Snowflake in workflows as well as agents?

Yes. Every Snowflake action also works as a step in a workflow, and the same connection serves your agents and your workflows.

Early access

Put Snowflake to work

Connect it in a few minutes, give an agent the actions it needs, and decide which changes wait for a person.