用 Jev 与 CodeGraph 组合做安全审查
开发者分享了 typesafe-security-review 项目:把 Jev 的概率判断与代码图、OWASP/CWE 信息结合,用于低成本安全扫描。
实验在5000词文章中隐藏50个提示注入,Jev和Clef全部捕获且误报不超过1次,而Horizon Guard误报46次。
But why would you even want decision model for injection detection when classifiers do the work? Additional experiment: hid 50 prompt injections in the middle of 5,000-word articles. Jev and Clef caught all 50 with ≤ 1 false flag. Horizon Guard also caught 50, but flagged 46/50