Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> I tend to think giving a remote party control over your command prompt inherently comes with risks.

I thought cursor (and probably most other) AI IDEs have this capability too? (source: I see cursor executing code via command line frequently in my day to day work).

I've always assumed the protection against this type of mishap is statistical improbability - i.e. it's not impossible for Cursor to delete your project/hard disk, it's just statistically improbable unless the prompt was unfortunately worded to coincidentally have a double meaning (with the second, unintended interpretation being a harmful/irreversible) or the IDE simply makes a mistake that leads to disaster, which is also possible but sufficiently improbable to justify the risk.





I only run ai tools in dev containers, so blast radius is somewhat minimal.

I don't think I've ever seen Claude even ask for permission for stuff outside of the directory it's working in

I run Codex in a sandbox locked to the directory it is working in.

umm, you have backups, right?



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: