What are Use Cases?
What are Use Cases?
Use Cases enable you to monitor LLM traffic across multiple Profiles simultaneously, designed specifically for complex agentic workflows where different interaction types require different security policies.
In traditional monitoring, you view traffic through a single Profile's security configuration. Use Cases allow you to define workflows where each event type (such as "User to LLM" or "Agent to Tool") can have its own Profile, and view all resulting events in a unified monitoring experience.
Why Use Cases?
Modern agentic applications involve multiple types of LLM interactions, each with different risk profiles and security requirements. For example:
User-to-agent interactions may need strict PII detection
Agent-to-tool calls may require different content filtering
Agent-to-LLM reasoning steps may have unique context validation needs
Use Cases let you define these differentiated security policies while maintaining visibility across your entire workflow.
How Use Cases Work
Configuring a Use Case
To create a new Use Case:
Save Requirements
You must configure either a default Profile or at least one event-type-specific Profile assignment to save the Use Case.
If you choose to use a combination of default and per-event-type Profiles, the default will apply to any event type not otherwise assigned.
After creation, you can edit the Use Case configuration from its detail view. To delete a Use Case, use the menu in the top right corner.
Monitoring Use Case Traffic
To view traffic from your agentic workflows:
What's Next
Use Cases are actively evolving. Upcoming enhancements include reporting and analytics across Use Case workflows.
This feature is a work in progress as we continue to build capabilities for agentic security and observability.
Last updated