longbridge-lbti

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill implements a standard quiz flow using local configuration files.
  • [EXTERNAL_DOWNLOADS]: The skill references image assets and QR codes hosted on the author's GitHub repository via a well-known CDN (jsDelivr). These resources (cdn.jsdelivr.net/gh/longbridge/extras-skills) are vendor-controlled and used for static content display.
  • [DATA_EXFILTRATION]: No evidence of unauthorized data access or exfiltration. The skill processes user answers within the chat context to determine a personality type and does not interact with sensitive files or external APIs for data submission.
  • [PROMPT_INJECTION]: The instructions describe a structured quiz logic where user input is limited to selecting predefined options (A/B). This input is used solely for logic branching to select a persona from local JSON files, preventing arbitrary instruction execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 08:37 AM
Security Audit — agent-trust-hub — longbridge-lbti