deviantintegral
Joined 5 years ago
Karma
234
Submissions
36
Member since
Mar 7
Joined 5 years ago
Reading "Claude Code instructs the receiving Claude never to change permission settings" and so on in the docs makes it seem like they're using LLM classifiers to determine what commands are safe or not.
I ended up going with a more expensive mini-PC style vendor, but I was surprised there didn't seem to be a ton of competitors.
Hopefully this situation encourages more contribution and improvement to tools like these.
I wish Apple (and Google, I presume) would actually enforce their app store guidelines and at least threaten to ban apps that do this. That seems like the only thing that'd actually have traction.
https://architecture.lullabot.com/adr/20211006-avoid-command...
I've been working on https://lullabot.github.io/sandbar/latest/ which works with Proxmox for VMs (and lima for locals or regular linux hosts over ssh). There's a diagram in https://lullabot.github.io/sandbar/latest/why/#recommended-w... with what we're currently recommending. Though, after some feedback, I'm in the process of integrating a colleague's web-based review tool as it turns out many preferred fully reviewing locally instead of using draft PRs.
It's got some opinions in terms of default tools for our team and industry so it may not fit yours. Forgive some of the AI-isms in the docs, I want to get the UX and feature set to a solid place before doing a full review.