trainer-design-quiz
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [Prompt Injection] (SAFE): The instructions define a specific persona ('experienced corporate trainer') and task scope without attempting to bypass safety filters or override global system instructions.
- [Indirect Prompt Injection] (INFO): The skill is designed to process external 'training content' provided by users. While this presents an ingestion surface, the skill has no dangerous capabilities (like code execution or network access) to exploit. The output is strictly formatted text.
- [Data Exposure & Exfiltration] (SAFE): No patterns for accessing sensitive local files (~/.ssh, .env, etc.) or making outbound network requests were detected. The mention of 'jinshuju.net' refers to a template format for manual import, not automated exfiltration.
- [Remote Code Execution / Dependencies] (SAFE): The skill consists entirely of markdown instructions. No package manifests (package.json, requirements.txt) or remote script downloads are present.
- [Obfuscation] (SAFE): Content is plain text markdown. No Base64, zero-width characters, or homoglyph-based evasion techniques were identified.
Audit Metadata