Comment by AlexCoventry
2 months ago
Run coding agents in a docker container with limited permissions. FWIW, I run it with
--cap-drop=ALL
--pids-limit=4096
--runtime=runsc
2 months ago
Run coding agents in a docker container with limited permissions. FWIW, I run it with
--cap-drop=ALL
--pids-limit=4096
--runtime=runsc
If you're on Linux, you can also easily run it in bwrap to properly sandbox without running a full container
Or put it in a microvm using eg smolmachines.
I've never used smolmachines but I'm curious; why this over a container?
Containers are not security boundaries. Vulnerabilities in containers are much more common than in VMs.
Kernel-level isolation is a significant security differentiator, for starters.
https://github.com/smol-machines/smolvm#comparison
shows a good comparison table.
Using runsc instead of runsc means that there's a hypervisor layer (gvisor, probably) in-between the kernel and the container userland
I run mine on their own machine, without root access.
Currently a Raspberry Pi 5
I am very pleased with it.
My Idiot Savant Pet