research-idea
Warn
Audited by Snyk on Jun 21, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.65). 该 skill 在运行时会读取用户提供的“任意资料/文件/URL/文本”并将其摘要/结构化事实写入隐藏工作区,再把这些内容作为输入喂给依赖的 LLM(如
research-topic-extractor、research-literature-review、parallel-vibe);若用户输入包含外部来源的自由文本(例如运行时抓取的网页内容、他人撰写的评论/issue/PR 讨论、下载附件正文),这些外部自由文本会进入 LLM 上下文。
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata