Snowflake
Audience: Admins, Developers, Solution Architects
Purpose: Explains what the Snowflake integration is, when to use it, and how it works within your Agentic Workflows
Overview
Run SQL queries and statements against your Snowflake warehouse directly from an Agentic Workflow in Kizen and use the results in later steps with the Snowflake integration.
What Is the Snowflake Integration?
The Snowflake integration adds two action steps to the Agentic Workflow builder:
Read Data: Runs a
SELECTquery against your Snowflake warehouse and returns the results to a variableWrite Data: Executes
INSERT,UPDATE, orDELETEstatements and returns the number of affected Records
You configure the connection to Snowflake once, as an integration secret in Kizen. See Enable Snowflake in Kizen for details.
The integration supports a single Snowflake environment or multiple environments, each with its own connection credentials. If you have more than one environment, you specify which to query when you call Read Data or Write Data from your Agentic Workflow.
To learn more, see Add Snowflake Data to an Agentic Workflow.
When to Use the Snowflake Integration
Use the Snowflake integration when an Agentic Workflow needs to check or update data that lives outside Kizen in a Snowflake warehouse before continuing to the next step. You don't need to copy the data into Kizen first. The integration handles the communication with Snowflake.
Common Use Cases
Surfacing a patient's utilization history across visits to flag care-gap risks before an appointment
Aggregating a policyholder's claims history to calculate a risk score during renewal review
Pulling a credit applicant's multi-year repayment and delinquency history to inform a portfolio risk model
How the Snowflake Integration Works
Before you configure the Snowflake integration, it's important to understand how it functions at a conceptual level.
Connection Model
The integration uses Snowflake SQL API version 2 over Hypertext Transfer Protocol Secure with Programmatic Access Token authentication to connect to Snowflake.
Connecting to Snowflake from Kizen requires a Snowflake user or service account and a role with the necessary USAGE privileges as well as read or write privileges on the tables to query. But instead of using an authorization protocol like OAuth, the Snowflake integration keeps track of your connection credentials as an integration secret. A single secret can hold credentials to multiple Snowflake environments, such as production and staging, as separate keys. You select which environment to use at each Read Data or Write Data step of the Agentic Workflow.
For more information, see Add Snowflake Data to an Agentic Workflow.
Query Execution
The integration doesn't use schema introspection. That is, it doesn't browse databases, schemas, or tables on your behalf. It runs the exact SQL statement you supply.
Queries and statements execute synchronously, so the Agentic Workflow waits at the Read Data or Write Data step until the database returns the result. The step times out on the Kizen side after 30 seconds.
Rate Limiting
Rate limits are subject to your Snowflake server's maximum number of connections, per-role connection limits, and timeout setting.
Results Handling
Both the Read Data and Write Data steps include the Return Single Value input, which, when true, signifies that you expect to return exactly one row and one column from the database. The result comes back as a string. When false, the full result set comes back as a single string in JSON format. To refer to the values of individual keys in later steps of the Agentic Workflow, parse the JSON string in a Code step.
Please also note the following:
Results sets larger than 5 MB might be truncated. Use
LIMITin the query to narrow the results.A result with multiple rows isn't auto-paginated. If you need pages, use
LIMITandOFFSETin the query.A query that returns no rows returns an empty string, not the null value.
Failure Handling
If a connection or query fails, for example because of invalid credentials, an unreachable database, or invalid SQL, the step fails immediately and results in an error message. Kizen doesn't automatically retry a failed step, so after you correct the underlying issue, rerun the step manually.
For more information, see Troubleshoot Snowflake.
Data Safety
If your Agentic Workflow writes data to your database, be sure to validate its behavior thoroughly on non-production data first. You can't roll back a Write Data operation after it executes, and the operation might partially apply if an error occurs during execution.
To maximize the safety of your data, grant the role assigned to the connected user or service account the appropriate level of access to Snowflake data and operations. If you don't intend to write data, for example, a read-only role with SELECT privileges on the specific tables to read is sufficient. Avoid roles with broad OWNERSHIP on Snowflake databases, schemas, and tables.
Available Snowflake Actions and Topics
For step-by-step instructions on using the Snowflake integration, see the following topics:
What's Next
Continue to Enable Snowflake in Kizen to configure Snowflake to work with Kizen, turn on the Snowflake integration, and enter your connection credentials as an integration secret.
Snowflake and all other Snowflake product, feature, and service names mentioned herein are registered trademarks or trademarks of Snowflake Inc. in the United States and other countries. Kizen's Snowflake integration is not affiliated with, sponsored by, or endorsed by Snowflake Inc.
Last updated
Was this helpful?