Snowflake Cortex Agents: How to use them in Sigma
Blog post from Sigma
Snowflake Cortex Agents let users ask plain-language questions of governed Snowflake data through semantic views and search services, while Sigma makes those agents available in workbooks, apps, and Sigma Assistant. Builders can connect one Cortex Agent to multiple Sigma agents, each with team-specific instructions, data sources, permissions, and actions; this allows answers to be combined with workbook data and turned into follow-up steps such as updating an input table or notifying colleagues. Snowflake roles and policies govern the data an agent can access, while Sigma adds controls over which users can use agents and what each Sigma agent can do; authentication determines whether access follows individual OAuth roles or a shared key-pair connection role. Setup involves creating a semantic view and Cortex Agent, granting the required Snowflake privileges, enabling and syncing the Snowflake connection in Sigma, granting Sigma catalog access, and adding the warehouse agent as a tool to a Sigma agent. Builders then specify which questions each source should answer and test that requests are routed correctly and unsupported questions are not answered with guesses. Sigma also supports other warehouse agents, including Databricks Genie Agents, so teams can select or change agents without rebuilding their workflows.
No tracked trend matches for this post yet.
Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.