Run bounded, evidence-driven training research through W&B Launch: assess project readiness, establish launchable code and queue capacity, smoke-test real jobs, execute serial trials, compare metrics, and persist resumable research state. Use when a coding agent is asked to autonomously test training hypotheses or tune a real W&B-tracked workload.
Install
npx skills add wandb/skills --skill wandb-autoresearchnpx skills add wandb/skills --skill wandb-autoresearch -a claude-code -gnpx skills add wandb/skills --skill wandb-autoresearch -a codex -ggemini skills install https://github.com/wandb/skills.git --path skills/wandb-autoresearchgh skill install wandb/skills wandb-autoresearchgit clone --depth 1 https://github.com/wandb/skills /tmp/wandb__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/wandb__skills/skills/wandb-autoresearch ~/.claude/skills/wandb-autoresearch # Claude Code
cp -R /tmp/wandb__skills/skills/wandb-autoresearch ~/.agents/skills/wandb-autoresearch # Codex, Cursor, Gemini CLI, Copilot, OpenCode