Trending storyPast story
Open-source prompt-injection detectors catch 0–1% of realistic AI agent attacks buried in tool output
1 report1 sourceupdated 6 days ago
What happened
Summary
这个基准测试把 629 条来自 AgentDojo 的注入攻击藏进搜索结果、邮件正文这类工具返回内容里，然后拿 Regex 和 Meta 的 Prompt Guard 2 去测。Regex 一条都没抓到，Prompt Guard 2 只抓到 1%。攻击不是直接发给模型的，而是混在模型调用的工具输出里，所以现有检测器基本是瞎的。代码和复现步骤都公开了，但...
Coverage
Follow the reports to see the story from different sides.
Sep 24
- Hacker News front pagePickOpen-source prompt-injection detectors catch 0–1% of realistic AI agent attacks buried in tool output
This benchmark hides 629 AgentDojo injection attacks inside tool outputs and tests Regex vs. Meta Prompt Guard 2. Regex catches 0%, Prompt Guard 2 catches 1%. The attacks aren't sent directly to the model—they're buried in search results, email bodies, and similar tool responses, so current detectors are effectively blind. Code and reproduction steps are public; the post doesn't include comparisons with commercial detectors.