crypto-frontend-design

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill explicitly "always gathers user context (images, theme, brand colors, audience)" and instructs using user-provided background images/asset URLs (e.g., "background: linear-gradient(...), url(...)") and partner logos, so it consumes and embeds arbitrary/untrusted third-party content (user-supplied images/URLs) as part of its generation workflow.