ask-questions-if-underspecified

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOW
Full Analysis
  • [PROMPT_INJECTION] (SAFE): The instructions do not attempt to bypass safety filters or override the agent's core identity. It reinforces responsible behavior by requiring explicit clarification.
  • [DATA_EXFILTRATION] (SAFE): No code or instructions for accessing sensitive files or transmitting data to external domains are present.
  • [REMOTE_CODE_EXECUTION] (SAFE): The skill contains no download commands (curl, wget) or execution patterns (eval, exec).
  • [COMMAND_EXECUTION] (SAFE): The skill explicitly commands the agent to avoid running commands or editing files until the user provides clarification.
  • [NO_CODE] (SAFE): This skill consists entirely of markdown-based instructions for reasoning; it does not include any scripts, dependencies, or executable binaries.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 09:14 PM