tooluniverse-variant-interpretation

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/REFERENCE.md explicitly instructs the agent to query and ingest public third‑party resources (e.g., ClinVar, gnomAD, PubMed/EuropePMC, bioRxiv/medRxiv, COSMIC, DisGeNET, PDB, UniProt, SpliceAI, AlphaFold) and to use those results to drive ACMG evidence codes and clinical recommendations, so untrusted external content is read and can materially influence decisions.