Reverse engineered prompt

Build me an Ansible based setup that can bring up and tear down a self hosted HA k3s Kubernetes cluster on a small group of Linux machines. I want it to work for Debian, Ubuntu, or Rocky nodes, and support both control plane and worker nodes, with embedded etcd when there is more than one master. It should set up a virtual API endpoint with kube vip, provide service load balancing with MetalLB, and make it easy to point kubectl at the finished cluster.

Please make the flow simple, copy the sample inventory into a real cluster folder, fill in hosts and shared settings like the API virtual IP and k3s version, then run one playbook to create the cluster and another to reset it. Include any collections or Python prerequisites, note the need for SSH access, and add a quick way to reboot nodes after teardown if needed. A short README with the exact commands and where to edit the inventory would be great. Look up current docs online if you need to.

Want more depth? Deep Reverse