scienceclaw-verification

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions define a rigorous workflow for decomposing scientific claims and validating them against evidence. No malicious patterns or security vulnerabilities were detected across all threat categories.
  • [COMMAND_EXECUTION]: There are no shell commands, scripts, or system-level operations defined in the skill content.
  • [DATA_EXFILTRATION]: No network operations or sensitive data access patterns are present. The skill references related research tools for legitimate data gathering purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:31 PM
Security Audit — agent-trust-hub — scienceclaw-verification