tooluniverse-drug-repurposing
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill focuses on medical and scientific data retrieval. The operations described, including querying specialized databases and analyzing protein/drug interactions, are consistent with the skill's legitimate purpose and do not involve unauthorized system access or data exfiltration.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8) as it ingests and processes content from external literature and databases like PubMed. However, the risk is mitigated by the domain-specific nature of the data and the lack of general-purpose code execution capabilities exposed to this data. (Ingestion points: PubMed_search_articles, EuropePMC_search_articles, drugbank_get_basic_info_by_drug_name_or_id in SKILL.md; Boundary markers: Absent; Capability inventory: Scientific analysis tools via ToolUniverse SDK; Sanitization: Absent)
Audit Metadata