questions-are-not-instructions

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [Prompt Injection] (SAFE): The skill provides behavioral guidelines for interaction. It does not contain instructions to bypass safety filters, ignore previous rules, or extract system prompts. Instead, it adds a layer of confirmation which can mitigate accidental execution of malformed user requests.
  • [Data Exposure & Exfiltration] (SAFE): There are no commands or patterns associated with file system access, credential harvesting, or network transmission.
  • [Remote Code Execution] (SAFE): The skill contains no code, scripts, or package manager commands.
  • [Obfuscation] (SAFE): No encoding, zero-width characters, or homoglyphs were detected. The content is entirely plain text.
  • [Persistence Mechanisms] (SAFE): No attempts to modify system configurations or shell profiles.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:40 PM