drug-discovery-informatics

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes external bioinformatics tools (AutoDock Vina and GNINA) via subprocess.run. These calls use list-based arguments, avoiding shell injection vulnerabilities, and are essential for the skill's primary purpose of molecular docking.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with the ChEMBL API via the chembl_webresource_client library to fetch molecular data. This is a well-known, trusted scientific service provided by the European Bioinformatics Institute (EMBL-EBI).
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill provides safety-oriented advice regarding domain-specific risks like PAINS filters and proper QSAR validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 08:29 PM