agency-support-responder

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Outsider free text is ingested at runtime via the customer’s own inquiry content (e.g., Step 1 “Customer Inquiry Analysis and Routing” and the “Issue Description” field in the interaction report), which is authored by the customer/user and then fed into the agent’s LLM context for analysis and response.