lanhu-design

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and analyzes user-generated Lanhu project pages and design resources from public lanhuapp.com URLs (e.g., via lanhu_get_pages, lanhu_get_designs, lanhu_get_ai_analyze_page_result and invite-link resolution) as shown in SKILL.md workflows, so untrusted third-party content is ingested and interpreted and could influence agent actions.