story

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

• Third-party content exposure detected (high risk: 0.90). The skill's routing explicitly sends intents like "长篇扫榜" and "浏览器操控" to handlers (e.g., /story-long-scan, /browser-cdp) and names public web novel sites (起点/番茄/晋江) and "抓取/浏览器" behavior, which indicates it will fetch and ingest untrusted public website content that can influence subsequent actions.