Playbook for launching, monitoring, stopping, and debugging NeMo-RL recipes on a Kubernetes cluster via the nrl-k8s CLI. Covers ephemeral vs long-lived RayCluster modes, iterating on runs, and debugging hung or failed training jobs.
Install
npx skills add NVIDIA/skills --skill launch-nemo-rlnpx skills add NVIDIA/skills --skill launch-nemo-rl -a claude-code -gnpx skills add NVIDIA/skills --skill launch-nemo-rl -a codex -ggemini skills install https://github.com/NVIDIA/skills.git --path skills/launch-nemo-rlgh skill install NVIDIA/skills launch-nemo-rlgit clone --depth 1 https://github.com/NVIDIA/skills /tmp/NVIDIA__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/NVIDIA__skills/skills/launch-nemo-rl ~/.claude/skills/launch-nemo-rl # Claude Code
cp -R /tmp/NVIDIA__skills/skills/launch-nemo-rl ~/.agents/skills/launch-nemo-rl # Codex, Cursor, Gemini CLI, Copilot, OpenCode