Nucleus: Enforced permissions for AI agents – policy+enforcement in one stack | HN Mirror

Y	Hacker News new \| ask \| show \| jobs

	Nucleus: Enforced permissions for AI agents – policy+enforcement in one stack (github.com)
	1 points by weitzj 30 days ago

1 comments

weitzj 30 days ago

There is also a Website https://coproduct-opensource.github.io/nucleus/architecture/...

I have stumbled on this project while dabbling around how to launch many firecracker VMs on a MacBook with >= M3 cpu to isolate agents.

What do others do to isolate agents?