customer-escalation

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow ingests the user-supplied “issue description” (and optional “customer name or account”) as free text, which can include outsider-authored content (e.g., customer-provided messages) and then the skill uses it to generate the escalation brief that is fed into the agent’s LLM context.