ui-agent-patterns

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

• Third-party content exposure detected (high risk: 0.90). The UIResearchAgent explicitly lists "search_web" (and methods like find_similar_components, analyze_competitors, gather_user_insights) and the skill references browser access, so the agent will fetch and ingest content from the open web/public sources and thus consume untrusted third‑party/user‑generated content as part of its workflow.