product-ui-taste
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external UI specifications and data samples to generate code, creating an attack surface for indirect prompt injection. Malicious instructions embedded in UI briefs could potentially influence the agent's behavior. Ingestion points: User-provided UI briefs and data shapes in SKILL.md. Boundary markers: None explicitly mentioned for untrusted content. Capability inventory: Generation of UI code and application shells. Sanitization: Not explicitly addressed.
- [EXTERNAL_DOWNLOADS]: The skill references and provides commands to install standard, well-known UI libraries and design systems from the official NPM registry, including IBM Carbon, Microsoft Fluent, and Shopify Polaris. These are reputable industry tools used for legitimate development purposes.
Audit Metadata