by anthropodie on 3/20/25, 6:40 PM with 17 comments
by roxolotl on 3/23/25, 3:59 AM
Of course you can evaluate everything it does before it does it. But no one reads the homebrew install script to make sure it’s safe when setting up a new Mac so who’s going to read the Claude code scripts?
by ics on 3/23/25, 3:16 AM
(No, I wouldn't seriously rm -rf ~/ intentionally in most cases but it's not so insane when you imagine creating a new user for project separation. For those reasons it's probably good to keep the agents separate too...)
by dorcy on 3/23/25, 3:51 AM
by water-data-dude on 3/24/25, 2:02 AM
by theogravity on 3/23/25, 4:22 AM
Best way to deal with this is to just clear the embedding index from the cursor settings and rebuild it.
I've never had it go to a point where it will want to rf home, but now I'm a bit fearful that one day it will go and do it as I have it on auto run currently.
by pcwelder on 3/23/25, 6:44 AM
Accidents will happen, disaster recovery is needed notwithstanding misaligned AI.
IMO once you're really ready for the worst, productivity from yolo mode AI assistance surpasses the downsides.
by casey2 on 3/23/25, 3:14 AM
Also Claude didn't "try" to do anything of course. Such a title is nonsense at best.