News for hackers. Stuff that matters.
Re:Goodbye Claude, Hello Pi by giladd on Sunday November 08, 2026 @01:05AM (#50410830)
Does Pi come with an auto mode that blocks dangerous actions? It didn't last time I checked, and these measures are critical to protect prod (and your personal files, etc) from a confused agent.

I gave the harness a try and it is refreshing, but the insistence on staying bear bones and not implementing basic functionality natively, together with no vetting process for popular 3rd party modules, made me return to Claude Code.

Re:Goodbye Claude, Hello Pi by JoshTriplett on Sunday November 08, 2026 @01:08AM (#50410840)
> bear bones

*bare

Re:Goodbye Claude, Hello Pi by giladd on Sunday November 08, 2026 @01:39AM (#50410962)
ty, edited
Re:Goodbye Claude, Hello Pi by bermudi on Sunday November 08, 2026 @01:09AM (#50410843)
You have time to watch over your agents?
Re:Goodbye Claude, Hello Pi by wmedrano on Sunday November 08, 2026 @01:10AM (#50410845)
You'd have to install an extension or write a function that intercepts tool calls.
Re:Goodbye Claude, Hello Pi by coldbrewed on Sunday November 08, 2026 @01:12AM (#50410858)
gotgenes/pi-permission-system and the ai-guard authorizer can do a lot here, it feels very similar to cc auto mode.

I built a policy authorizer on top of pi-permission-system that looks kinda like nftables that models commands/subcommands/options as read/write and enfores policies depending on if the sub agent is read only or write, which really cuts down on the stupid decisions and nicely complements AI cmd review. Pi shines for making things like that possible.

Re:Goodbye Claude, Hello Pi by jagrit on Sunday November 08, 2026 @01:25AM (#50410905)
like the auto mode classifier? if it doesn't then... that's a risky tradeoff
Re:Goodbye Claude, Hello Pi by agentdev001 on Sunday November 08, 2026 @01:32AM (#50410938)
Sounds to me like it isnt basic functionality if it all works ;)
Re:Goodbye Claude, Hello Pi by pkulak on Sunday November 08, 2026 @01:37AM (#50410956)
Well yeah, the whole point of pi is that you build it to your workflow. If you dislike the entire point of a project, it’s not for you. I run my agents in a sandbox. When Claude or Codex then double sandbox themselves, it gets really annoying.
Re:Goodbye Claude, Hello Pi by mpern on Sunday November 08, 2026 @03:14AM (#50411345)
I would not trust auto mode in the first place.

Wrap your agent of choice in e.g. docker sandbox, nono.sh, Nvidia OpenShell, microsanbox or any other sandboxing tool and run it without any checks (i.e. "yolo mode", dangerously-skip-permissions etc.)

IMHO it makes the whole category of problems around permission fatigue, trusting the harness (or plugin) authors etc go away.

Re:Goodbye Claude, Hello Pi by giladd on Sunday November 08, 2026 @05:10AM (#50411897)
I did that. Customized Pi to run in Bubblewrap, made an action authorization flow, etc. But that's throwing the baby with the bathwater. This gives you a harmless, less capable agent. To get both safety and usefulness, Ant's solution of a fast classifier works really well.
Re:Goodbye Claude, Hello Pi by mpern on Sunday November 08, 2026 @07:23AM (#50412660)
For me, its the other way around.

Strong sandbox + disabling permission checks gives me maximum capabilities, least annoyance and (relative) peace of mind that any prompt injection, malicious plugin,... won't wreck my shit or leak credentials of any kind

Re:Goodbye Claude, Hello Pi by owlmusic on Sunday November 08, 2026 @03:44AM (#50411488)
I use https://pi.dev/packages/pi-warden with Jev (but can use other Decisions API), it would make sense to combine with pi-permission-system to get the CC two stage "default auto" mode.