Periodically check WandB metrics during training to catch problems early (NaN, loss divergence, idle GPUs). Avoids wasting GPU hours on broken runs. Use when training is running and you want automated health checks.
Install
npx skills add wanshuiyin/Auto-claude-code-research-in-sleep --skill training-checknpx skills add wanshuiyin/Auto-claude-code-research-in-sleep --skill training-check -a claude-code -gnpx skills add wanshuiyin/Auto-claude-code-research-in-sleep --skill training-check -a codex -ggemini skills install https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git --path skills/training-checkgh skill install wanshuiyin/Auto-claude-code-research-in-sleep training-checkgit clone --depth 1 https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep /tmp/wanshuiyin__Auto-claude-code-research-in-sleep
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/wanshuiyin__Auto-claude-code-research-in-sleep/skills/training-check ~/.claude/skills/training-check # Claude Code
cp -R /tmp/wanshuiyin__Auto-claude-code-research-in-sleep/skills/training-check ~/.agents/skills/training-check # Codex, Cursor, Gemini CLI, Copilot, OpenCode