tooluniverse-immune-repertoire-analysis

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

• Third-party content exposure detected (high risk: 0.80). The skill's Phase 7 (SKILL.md) and the functions query_epitope_database / predict_specificity_vdjdb in ANALYSIS_DETAILS.md explicitly call external services (IEDB_search_tcells, PubMed_search and reference VDJdb at vdjdb.cdr3.net) to fetch public third‑party data which the agent parses and uses to drive epitope-specificity predictions and downstream decisions, exposing it to untrusted external content that could influence behavior.