bilibili

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

• Third-party content exposure detected (medium risk: 0.30). When the required workflow runs the bilibili skill, it calls Bilibili read endpoints (e.g., GET https://api.bilibili.com/x/space/wbi/article and GET https://api.bilibili.com/x/article/viewinfo) and ingests the returned article titles/stats/reasons text (which are outsider-authored by other users who wrote the articles).