metabolomics

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides standard scientific instructions for metabolomics workflows using established tools like pandas and seaborn. No indicators of prompt injection, data exfiltration, or malicious intent were found.
  • [COMMAND_EXECUTION]: The skill includes instructions to verify software versions using python -c and <tool> --version. These are standard practices for environment validation and do not pose a security risk in this context.
  • [DATA_EXPOSURE_EXFILTRATION]: The workflow involves reading and writing data (e.g., feature_table.csv, results/, figures/). These operations are local and aligned with the stated scientific purpose of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 03:53 AM
Security Audit — agent-trust-hub — metabolomics