hypothesis-generation

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

• Third-party content exposure detected (high risk: 0.80). This skill instructs the agent to fetch and read open web literature using "WebFetch" for PubMed URLs and "WebSearch" for general web sources (including preprint servers like arXiv/bioRxiv and other public websites), exposing it to untrusted, user-published third-party content that the agent must interpret as part of its workflow.