environment

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides technical documentation and usage examples for built-in platform tools such as execute_code, Bash, and ouija. It contains no executable scripts or malicious code.
  • [SAFE]: The guide highlights security-conscious platform constraints, specifically noting that destructive git operations (push --force, reset --hard) are blocked by default.
  • [SAFE]: Instructions regarding context management and the use of supervisor controls are presented as performance and resource optimizations rather than attempts to evade safety protocols.
  • [SAFE]: No obfuscation, data exfiltration patterns, or unauthorized remote code references were detected within the guide content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:23 PM
Security Audit — agent-trust-hub — environment