inspired-product

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

• Third-party content exposure detected (medium risk: 0.65). Outsider free text is ingested when the skill’s runtime workflow includes customer interviews/user testing (e.g., “interviewed 15 recently churned customers” and “quote customers verbatim”), where participants’ spoken/written responses are transcribed and fed into the agent’s LLM context.