vida-perspective

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides roleplay instructions for a specific persona. It does not contain any malicious prompt injection patterns aimed at bypassing AI safety guardrails; rather, it explicitly instructs the agent to refuse requests for illegal or fraudulent operations.
  • [SAFE]: No network operations, such as 'curl' or 'wget', or data exfiltration patterns were found. The skill relies entirely on a provided local dataset.
  • [SAFE]: No hardcoded credentials, API keys, or access to sensitive local file paths (like SSH keys or environment files) are present.
  • [SAFE]: The skill does not include any external dependencies, package installation manifests, or remote code execution vectors.
  • [SAFE]: No obfuscation techniques or hidden instructions were identified within the markdown or JSON files.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 11:16 AM