taste-audit

Pass

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: No external downloads, dependencies, or remote resources are referenced in the skill content.
  • [DATA_EXFILTRATION]: The skill does not possess capabilities to read sensitive files or perform network operations, precluding data exfiltration risks.
  • [PROMPT_INJECTION]: No malicious instructions aimed at bypassing agent safety guidelines or overriding system prompts were detected.
  • [REMOTE_CODE_EXECUTION]: The skill is purely text-based and does not include any scripts, commands, or logic for remote or dynamic code execution.
  • [COMMAND_EXECUTION]: No shell commands or subprocess calls are utilized within the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 24, 2026, 05:07 AM