tooluniverse-gwas-study-explorer

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements specific bioinformatics analysis logic using a controlled tool-calling interface (tu.run). It does not contain any patterns for unauthorized network access, data exfiltration, or arbitrary command execution.- [SAFE]: No obfuscation, prompt injection, or privilege escalation attempts were detected within the skill's instructions or Python scripts. Documentation and code align with the stated scientific purpose.- [SAFE]: The skill demonstrates secure secret management by providing an environment variable template with placeholders instead of hardcoded API keys.- [SAFE]: Data ingested from genomic APIs is processed using standard parsing techniques (regex for numeric values and string comparisons) that do not present a significant surface for indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 11:48 PM