research-plan
Warn
Audited by Snyk on Jun 18, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). 该 skill 的运行时流程包含“文献检索/下载 PDF 并从 PDF 提取文本(阶段 1.2-1.5、阶段 2.1)”,这些步骤会读取来自外部来源(如 PubMed/Google Scholar/Semantic Scholar/arXiv/Unpaywall/任意开放获取链接)的论文正文/元数据文本进入 LLM 上下文,属于“公共 web 内容/外部文献文本”导致的间接提示注入风险。
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata