ui-design-review

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly requires "screenshots_or_urls" as a required input (visual references of the interface), which means the agent is expected to ingest and interpret user-provided or public webpage content (arbitrary URLs or screenshots) and therefore can be exposed to untrusted third-party content that could carry indirect prompt injection.