story-long-scan

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

• Third-party content exposure detected (high risk: 1.00). The skill's workflow and scripts explicitly use browser‑cdp and web‑fetch scrapers to crawl public, user‑generated ranking pages (e.g., qidian.com, fanqienovel.com, qimao.com, jjwxc.net, ciweimao.com) and then parse and analyze that content as input to its reports and recommendations, so untrusted third‑party content can materially influence agent decisions.