deepsearch

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE] (SAFE): The skill contains no executable logic, minimizing the attack surface to zero for traditional code-based threats.
  • [NO_CODE] (SAFE): No Python, Node.js, or shell scripts are present; the skill is entirely comprised of markdown documentation.
  • [PROMPT_INJECTION] (SAFE): The instructions follow a clear, task-oriented structure without attempting to override system prompts or safety constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:18 PM