Snowflake Workbench

A Streamlit-based Snowflake client that behaves like a lightweight DBeaver-style workbench:

• Saves reusable Snowflake connection profiles
• Connects to Snowflake with a PAT token
• Browses databases, schemas, tables, and views
• Runs multi-statement SQL
• Shows result grids and query history
• Describes objects, previews table data, and fetches DDL
• Exports result sets to CSV

Why PAT auth is implemented this way

Snowflake's current documentation shows that a programmatic access token is used as the password when connecting with the Python connector. This app follows that model.

Run locally

1. Create and activate a virtual environment.

2. Install dependencies:

   pip install -r requirements.txt

3. Start the app:

   streamlit run app.py

4. In the sidebar, enter:

• Click New Snowflake Connection
• Save the Snowflake account identifier, user, PAT token, and optional warehouse/role/database/schema/host
• Select the saved connection from the connection dropdown
• Use the database, schema, and table selectors in the explorer

Optional environment variables

You can prefill the connection form with:

• SNOWFLAKE_ACCOUNT
• SNOWFLAKE_USER
• SNOWFLAKE_WAREHOUSE
• SNOWFLAKE_ROLE
• SNOWFLAKE_DATABASE
• SNOWFLAKE_SCHEMA
• SNOWFLAKE_HOST

Notes

• PAT support depends on Snowflake account-side enablement and token issuance.
• Saved connection profiles are stored locally in .streamlit/snowflake_connections.json.
• The app intentionally keeps the UI focused on query workbench flows rather than cloning every DBeaver feature.
• The PAT input is masked in the modal. Saved profiles include the PAT in the local JSON file so the profile can reconnect without re-entry.