Skip to content

Database Access

Configuring Access

In Model Options → Databases, click Add Database to give your Assistant Model structured access to your data.

Sidebar – Database Access

Options Explained

  • Name – A friendly label (e.g., “Sales Warehouse”).
  • Connection String – Defines how Assistant Engine connects (server, database, credentials).
    Your database administrator can supply this.
  • Dialect – The database type (e.g., SQL Server, PostgreSQL, MySQL).
  • Describe Database – If enabled, Assistant Engine uses the Descriptor Model to generate summaries of your schema before ingestion.
    This improves semantic search and SQL generation accuracy.

Database Considerations

  • Permissions – Ensure the user account in your connection string has read access at minimum.
  • Schema Size – Large schemas may take longer to ingest and vectorize.
  • Updates – Re-ingest whenever your schema changes to keep knowledge up to date.
  • Security – Store credentials securely; never commit them to source control or share in plaintext.

Best practice

After adding a database:
1. Test Connection to confirm credentials and connectivity.
2. Run Ingest to build the vector store.
3. Re-ingest if your schema changes.

Built-In Database Features

Once configured, Assistant Engine provides several database-aware tools:

  • ExecuteSQL – Run direct SQL queries by asking the Assistant in natural language.
  • Search Schema – Retrieve tables, columns, and relationships from your ingested schema.
  • Search Database – Perform semantic search across stored schema data for context-aware answers.

When to use ExecuteSQL

Use this feature carefully — the Assistant will execute queries directly against your database.
Best suited for read-only accounts or sandbox environments to avoid accidental data changes.