scienceclaw-retrieval

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidelines for scientific database searches without containing executable scripts or dangerous commands.
  • [SAFE]: No credentials, sensitive file access, or unauthorized network operations were detected.
  • [SAFE]: The search strategies and database-specific query examples (PubMed, arXiv, etc.) follow standard academic research methodologies.
  • [SAFE]: The skill lacks obfuscation and does not include any external dependencies or remote code execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:31 PM
Security Audit — agent-trust-hub — scienceclaw-retrieval