connecting-streamlit-to-snowflake

Connect Streamlit to Snowflake securely using st.connection and st.secrets to manage credentials in a secrets.toml file, avoiding hard-coded credentials and enabling automatic connection pooling. This pattern keeps your database access safe and performant.

Prevent SQL injection in Streamlit Snowflake queries by using parameterized queries. Passing user inputs as parameters rather than interpolating strings ensures the database treats them strictly as data, neutralizing malicious SQL code.

Caching for Snowflake queries in Streamlit works by using TTL-based caching to store query results. This reduces query load and latency by reusing cached data for subsequent requests within the defined time-to-live period, improving app performance.

Yes, you can enforce viewer-based row access in a Streamlit Snowflake app by using caller's rights. This permissions model ensures the session executes with the viewer's specific role, displaying different rows to different users based on their access level.

Yes, Streamlit works with Snowflake Cortex LLMs by using a Snowflake session to stream LLM completions. This integration enables you to build interactive chat UIs that generate streaming assistant responses directly from your database session.