smina-molecular-docking

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess.run to execute external scientific tools. It correctly uses argument lists rather than shell strings, which effectively prevents command injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: Dependencies such as smina, rdkit, and prody are sourced from well-known and trusted package managers (conda-forge, PyPI) and official academic repositories (The Scripps Research Institute).
  • [DATA_EXFILTRATION]: No sensitive data access or unauthorized exfiltration was detected. Network operations are limited to fetching publicly available molecular structures from the Protein Data Bank (PDB).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:35 PM
Security Audit — agent-trust-hub — smina-molecular-docking