article-analyzer

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

• Third-party content exposure detected (high risk: 0.80). The skill's "联网检索规范" explicitly requires performing online fact-checks and background/expanded searches and to "标注来源", which means the agent will fetch and read open/public third‑party web content (arbitrary websites/social sources) as part of its workflow, exposing it to untrusted user-generated content and potential indirect prompt injection.