I’m glad to find other people who actually care enough to sandbox their agents. I don’t have a kanban board, but I also built a multi-agent manager on top of gvisor (also kubernetes, but still runs on a single machine with kind). Recently switched to git reference clones instead of worktrees for better security. The transparent egress proxy + secure credential isolation was also kind of a pain to figure out. Thought I’d link here in case any of the code is useful to you, or if you want to try it out: https://yaac.dev/