tooluniverse-cancer-variant-interpretation
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not contain any detected malicious patterns, obfuscation, or unauthorized access to sensitive data.- [EXTERNAL_DOWNLOADS]: The skill performs network operations to retrieve biological and clinical data from reputable scientific sources including MyGene, UniProt, OpenTargets, CIViC, cBioPortal, FDA, DrugBank, ChEMBL, PubMed, Reactome, and GTEx. These operations are essential for the skill's stated purpose and target well-known research infrastructure.- [DATA_EXFILTRATION]: There is no evidence of sensitive local file access or unauthorized data exfiltration. The skill processes clinical data to generate user-requested reports.- [PROMPT_INJECTION]: The skill includes instructions to ingest and process data from external research databases. While this provides a surface for indirect prompt injection, the risk is minimal as the skill operates within a specific scientific domain and does not possess capabilities to execute arbitrary system commands based on that data.
- Ingestion points: External databases (CIViC, PubMed, etc.) and user-provided gene/variant strings.
- Boundary markers: The workflow follows a structured sequence of analysis phases.
- Capability inventory: Network operations to verified scientific APIs; file creation for markdown reports.
- Sanitization: Not explicitly documented, but the data is processed into structured report sections.
Audit Metadata