ux-audit

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

• Third-party content exposure detected (medium risk: 0.30). The skill ingests outsider-authored free text at runtime via WebFetch on user-provided brand.caseStudyUrl to extract problem/features/quotes for the report (Phase 1 step 5), and via web-scraping brand.portfolioUrl HTML to source cover/logo images (Phase 1 step 5), before generating narrative content.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The HTML checklist/template includes a runtime script tag loading and executing https://mcp.figma.com/mcp/html-to-design/capture.js (references/day-1-audit-checklist.html:7-8), which the skill's HTML→Figma capture fallback relies on at runtime to perform remote capture actions, so this is an external runtime dependency that executes remote code.