tooluniverse-adverse-event-detection

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests public, third-party safety data (notably FAERS via FAERS_count_*/FAERS_search_adverse_event_reports and FAERS_calculate_disproportionality in Phases 1–2) and external literature/preprints (PubMed, EuropePMC, OpenAlex in Phase 7), and it uses that untrusted/user-submitted content to compute disproportionality metrics and the Safety Signal Score that directly influence decisions and recommendations—creating a clear avenue for indirect prompt injection.