Gezinmeyi atla
The next prompt injection your coding agent reads might be an error report.
0:00 / 21:19

The next prompt injection your coding agent reads might be an error report.

Cloudflare Developers

@CloudflareDev86 B takipçi

X profili
196 beğeniİndirOrijinal gönderi

15.328 görüntüleme

The next prompt injection your coding agent reads might be an error report. So Sentry screens for it first, spending $263 a day on Jev to check 10M events. Now the team is testing its own classifier, aiming to match Jev's accuracy for 90% less. Greg Pstrucha explains how.

10 Yorum

Yorumlar X (Twitter) üzerindeki orijinal gönderiden alınmıştır

  • Anand Butani profil fotoğrafı

    @AnandButani

    @grichadev building an in-house classifier to save $230 a day is peak infra behavior

    2 beğeni

  • Greg Pstrucha profil fotoğrafı

    @grichadev

    boy i need some beta blockers next time

    1 beğeni

  • יוסי profil fotoğrafı

    @random11guy

    @grichadev For 2000$ per month? What

  • Marlon Steiner profil fotoğrafı

    @MarlonCEO

    @grichadev Stack traces as an attack surface is the kind of thing nobody writes down until it bites someone. Does the classifier sit in front of the agent, or does the agent read the raw event either way?

  • TheAsymmetricMind profil fotoğrafı

    @asymmetricmind

    @grichadev Screening error reports before the agent reads them is smart, and it will always run a step behind. Any text an agent reads can steer it, and a classifier is one more model guessing. What has to hold is the action: a check on every tool call, outside the model, default no.

  • HM profil fotoğrafı

    @a0d0c0

    @grichadev Error reports are a sneaky place for this. I'd want the original event kept intact for debugging, with the suspicious bits clearly marked before the agent reads it.

  • Pyra profil fotoğrafı

    @Pyraio

    @grichadev At 10M events a day, matching Jev's accuracy is only half the job. How are you planning to catch drift when attackers start shaping error reports around the in-house classifier?

  • Rotex Hawk 🦇🔊 profil fotoğrafı

    @Rotexhawk

    @grichadev Why not use clef?

  • Lane Henslee profil fotoğrafı

    @LaneHenslee

    @grichadev The attacker's new job is writing an error report that the cheap classifier calls safe and the coding agent still obeys

  • red_square profil fotoğrafı

    @red___square

    @grichadev Thought you had clef!?