tooluniverse

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests public third-party data (e.g., GEO_download_dataset, PubMed_search, OpenTargets_get_associated_targets_by_disease_efoId, UniProt_get_sequence, PDB/PubChem/ChEMBL access shown in references and example scripts), and then processes/summarizes that content (e.g., LLM_summarize, workflow steps), exposing the agent to untrusted user-generated or public web content that could enable indirect prompt injection.