tooluniverse-immunotherapy-response-prediction

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md) explicitly calls numerous public third‑party APIs (e.g., OpenTargets_get_associated_drugs_by_disease_efoId, PubMed_search_articles, clinical_trials_search, FDA_get_indications_by_drug_name, civic_search_evidence_items, cBioPortal_get_mutations, IEDB, DrugBank, HPA, gnomad, etc.) and ingests their results to drive scoring and treatment recommendations, so external (public) content can materially influence tool use and next actions.