technical-pm

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow repeatedly instructs reading and ingesting external literature (e.g., "PubMed search", "Review hollow fiber literature", Skill(researcher, topic=...)", and expects the agent to read researcher outputs/scratchpad and use those findings to drive calculations, model choices, and design recommendations—i.e., it fetches and interprets public third‑party content that can materially influence downstream actions.