Posts by cibyr@omg.wtf.sh
 (DIR) Post #B8fnl81tNpN5KCsefY by cibyr@omg.wtf.sh
       0 likes, 0 repeats
       
       @ariadne "run the agent in a sandbox so that the dumb thing inside accidentally rm -rf's my box less frequently" and "run the agent in a sandbox so that the superhuman hacker inside can't break out and take over my whole network" are two VERY different requirements and people are acting as if the latter is just a slightly harder version of the former
       
 (DIR) Post #B8foU1yyiFFDbMvddw by cibyr@omg.wtf.sh
       0 likes, 0 repeats
       
       @ariadne a lot of engineering is embracing having "already lost". Gotta have backups anyway, so question is "is having to clean up the AI's mistakes worse than having to do this task myself?" A shoddy, weak sandbox might be useless as an actual security control while still making a meaningful difference to that trade-off
       
 (DIR) Post #B8ggFwGeYi5tLcvQci by cibyr@omg.wtf.sh
       0 likes, 0 repeats
       
       @ariadne @poliorcetics that's...what I did? It's not like there's a button in home assistant for "automatically find the right threshold for a bathroom fan and this humidity sensor". I could point-and-click through the automation UI, or write some yaml (probably with a static threshold), or try to find something someone else has built (hope they have the same assumptions and preferences as me!), or sit down and learn how to write a home assistant integration, or ..just ask claude to do it. WCGW?
       
 (DIR) Post #B9oM7Gc5s1MbxxZKHQ by cibyr@omg.wtf.sh
       0 likes, 0 repeats
       
       @ariadne I've used it, and liked it. Been running in production for years now with, AFAIK, no drama. Friendly, responsive maintainers too