bioagents-guide

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly describes a "Literature" module that ingests public sources (PubMed, Semantic Scholar, bioRxiv) and the generate_hypotheses workflow takes literature_context as input which the agent is expected to read and use to drive hypotheses and experimental designs (including downstream lab submission), so untrusted third‑party content can materially influence actions.