tooluniverse-immunology
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill processes data from various external scientific databases based on user-provided queries, creating a surface for indirect prompt injection.
- Ingestion points: External data enters the agent's context through tools defined in SKILL.md that query databases such as IEDB, GWAS, Orphanet, and KEGG.
- Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore instructions within the retrieved database content.
- Capability inventory: The skill (SKILL.md) enables the agent to fetch structural, genetic, and clinical data and perform enrichment analysis.
- Sanitization: Absent; there are no defined processes for validating or escaping the content returned from external scientific services.- [EXTERNAL_DOWNLOADS]: The tool SAbDab_get_structure provides information and download links for antibody structures (PDB files) hosted on official structural biology repositories.
Audit Metadata