foldseek-structural-search

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate bioinformatic analysis by interacting with the official Foldseek web API (search.foldseek.com). All network operations are scoped to this well-known scientific service.\n- [SAFE]: The Python script implements strict validation of user inputs, ensuring that only specific 3D coordinate file types (.pdb, .cif, .mmcif) are processed and only authorized databases are queried.\n- [SAFE]: Sensitive data access is restricted to the specific protein structure files provided by the user for the purpose of the search. No evidence of credential harvesting or access to unauthorized system paths was found.\n- [SAFE]: The skill uses standard dependency management via uv and relies on a local utility (science-skills-common) for managed HTTP requests, which is a best practice for agent tools.\n- [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation patterns were detected in the instructions or the accompanying script.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 06:50 PM
Security Audit — agent-trust-hub — foldseek-structural-search