quiz

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSPRIVILEGE_ESCALATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill renders user-provided JSON content into HTML using innerHTML. This represents a potential XSS vulnerability if the input data is malicious, although it is consistent with the skill's purpose as a frontend renderer.\n
  • Ingestion points: Loads data from a JSON file via load_quiz_data in main.py.\n
  • Boundary markers: None identified; data is directly interpolated into the output template.\n
  • Capability inventory: The skill performs file writes to create the resulting HTML quiz file.\n
  • Sanitization: Employs json.dumps to embed data in script blocks, but renders text content through innerHTML.\n- [EXTERNAL_DOWNLOADS]: The generated HTML fetches KaTeX libraries from JSDelivr, a well-known and trusted CDN service.\n- [PRIVILEGE_ESCALATION]: The scripts/install.sh file utilizes the --break-system-packages flag for pip. This is a utility pattern used to install dependencies in environments that restrict global package modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 11:31 PM
Security Audit — agent-trust-hub — quiz