emil-design-eng

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The FIELD-GUIDE.md file contains directives to control the agent's interaction flow, such as requiring a specific greeting upon first invocation ("> I'm ready to help you...") and commanding the agent to "Do not provide any other information until the user asks a question." These are functional constraints for the skill's persona rather than adversarial bypass attempts.
  • [PROMPT_INJECTION]: The skill enforces strict output formatting rules, mandating the use of markdown tables for UI reviews and explicitly forbidding list formats ("MUST use a markdown table... Do NOT use a list... Wrong format (never do this)").
  • [EXTERNAL_DOWNLOADS]: The documentation references legitimate external design tools and educational resources, including animations.dev, easing.dev, and easings.co. These are well-known services in the web development and design community.
  • [SAFE]: The skill references standard front-end libraries like framer-motion (Motion) and Sonner for instructional purposes. No commands for automated installation or execution of untrusted third-party code were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 05:45 AM
Security Audit — agent-trust-hub — emil-design-eng