Structural Cloud LLM configuration

Models used on Structural Cloud

For its AI-enhanced features, Structural Cloud uses Opus 5, Sonnet 5, and Haiku 4.5 from Anthropic deployed on Amazon Bedrock.

Configuring AI-based Structural functionality

Opting out of all Structural AI features

Cloud organizations can choose not to use any Structural AI features.

On the Organization Settings page of Structural Settings, toggle TONIC_LLM_PROVIDER to the off position.

Excluding sample data from LLM-based sensitivity detection

By default, LLM-based sensitivity detection sends both the database schema and sample source data values to the LLM.

To exclude the contextual source data, and only send the schema, on the Organization Settings tab of Structural Settings, set TONIC_LLM_ENABLE_ENHANCED_RECOMMENDATIONS_SAMPLE_DATA to false.

Disabling LLM-based sensitivity detection in a workspace

On Structural Cloud, LLM-based sensitivity detection is enabled by default.

You can disable LLM-based sensitivity detection within individual workspaces.

To do this, in the advanced workspace overrides, set TONIC_LLM_ENABLE_ENHANCED_RECOMMENDATIONS to false.

LLM-based sensitivity detection also processes LLM-based sensitivity rules. If you disable LLM-based sensitivity detection, then the sensitivity scan does not process any LLM-based sensitivity rules.

Excluding sample data from all LLM-based Structural features

To configure Structural to not send sample data for any LLM-based features at all, on the Organization Settings tab of Structural Settings, set TONIC_LLM_ENABLE_SAMPLE_DATA to false.