TGViewer
Tech RSS Tech RSS @techrss · 707 subscribers
Post #399473 51
 Can open-source prompt-injection detectors catch realistic AI agent attacks?

https://github.com/rudratoshs/buried-injections
GitHub GitHub - rudratoshs/buried-injections: 🛡️ Regex catches 0%, Meta's Prompt Guard 2 catches 1% of 629 realistic AgentDojo injection… 🛡️ Regex catches 0%, Meta's Prompt Guard 2 catches 1% of 629 realistic AgentDojo injection attacks when they're buried in tool output. Reproducible benchmark. - rudratoshs/buried-injections
Open in Telegram All posts from @techrss
More from @techrss
  1. Oct 14, 2026How the Military Will Hack the Election https://www.kenklippenstein.com/p/how-the-military…
  2. Oct 14, 2026Fire at Starlink site in Bulgaria after suspected sabotage in Poland https://dpa-internati…
  3. Oct 14, 2026Banks are now trying to offload the AI data center debt they just underwrote https://start…
  4. Oct 14, 2026Show HN: Petal, open-source CoScreen/ScreenHero finally free https://petal.live/
  5. Oct 14, 2026America Didn't Lose Its Rare Earth Industry. It Sold It https://www.lawfaremedia.org/artic…
  6. Oct 14, 2026Neckbeard – a DevOps/platform/SRE plugin for your agent https://github.com/bealesh/neckbea…
Threads Profile ViewerView any public Threads profile without an account.Open ThreadLook →Writing with AI? Make it sound human.Metric37 rewrites AI drafts so they read naturally. Free AI detector, 1,500 words free.Try Metric37 →