Tests prompt injection in tickets, attachments, and retrieved content, along with cross-customer leakage, policy hallucination, and refund abuse.
Scopes CRM, order, and refund tools to the customer's request and escalates sensitive actions.
Customer support agents
Support agents search knowledge bases, read account history, update tickets, issue refunds, and communicate directly with customers. One manipulated message can turn legitimate access into data leakage or fraud.
Tests prompt injection in tickets, attachments, and retrieved content, along with cross-customer leakage, policy hallucination, and refund abuse.
Scopes CRM, order, and refund tools to the customer's request and escalates sensitive actions.
Support agents search knowledge bases, read account history, update tickets, issue refunds, and communicate directly with customers. One manipulated message can turn legitimate access into data leakage or fraud.
Tests prompt injection in tickets, attachments, and retrieved content, along with cross-customer leakage, policy hallucination, and refund abuse.
Scopes CRM, order, and refund tools to the customer's request and escalates sensitive actions.
A customer asks why an order is delayed. An attached document instructs the agent to export all customer records. dFence allows the order lookup but blocks the unrelated export because its intent falls outside the support workflow.
For every agent, dwaar asks four questions: What is the agent supposed to do? Which data and tools should it reach? How could that workflow be manipulated? What should happen when its behavior drifts?
Document the happy path, permitted tools, data boundaries, and approval points.
dForge attacks the workflow before launch using context-specific prompt, tool-use, and multi-turn tests.
dFence evaluates runtime intent before sensitive tool and MCP calls, then allows, blocks, constrains, redacts, or requests approval.
Turn verified findings and runtime events into stronger policies and repeatable security tests.
Choose a use case to see its risks, adversarial tests, runtime controls, and security decisions in context.
dwaar combines agent-specific adversarial testing with intent-aware runtime enforcement, helping teams deploy capable agents without relying on static scopes alone.