tooluniverse-antibody-engineering

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow calls outsider-authored free text via public database/tool outputs (e.g., TheraSAbDab_search_by_target, SAbDab_search_structures, IMGT_get_sequence, iedb_search_epitopes), which return readable records/annotations from external sources that the agent then incorporates into the LLM context for report generation.