design-taste-frontend
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a high-quality instructional framework for generating modern frontend interfaces. It imposes strict aesthetic and technical constraints to ensure the output remains professional and avoids common AI-generated design patterns.
- [EXTERNAL_DOWNLOADS]: The skill references numerous external design systems and utility libraries (e.g., Tailwind CSS, Radix UI, Microsoft Fluent UI, IBM Carbon, GSAP, Framer Motion). These references are directed toward official GitHub repositories, package registries (NPM), and well-known documentation domains (microsoft.design, carbondesignsystem.com, shopify.dev, etc.). These are trusted sources for development resources.
- [COMMAND_EXECUTION]: The skill provides standard installation commands using
npmandnpxfor various design systems. These are intended for the user's project setup and do not involve the execution of unknown or untrusted scripts. - [DATA_EXFILTRATION]: There is no evidence of sensitive data access or unauthorized network transmissions. The skill suggests using established external assets for placeholder images (
picsum.photos) and brand icons (simpleicons.org), which are standard industry practices. - [PROMPT_INJECTION]: The skill includes extensive internal 'Pre-Flight Checks' and 'Anti-Slop' rules. These are designed to constrain the agent's output for quality control rather than overriding system-level safety protocols.
Audit Metadata