Authors a new Instance AI workflow or Agent eval case — written locally as JSON, calibrated against a real build, then pushed to the LangTracer suite CI runs — build cases, behaviour/process cases, credential cases, and seeded (mid-conversation) cases — with intent-driven expectations. Use when adding or changing an Instance AI eval, or debugging why one is flaky.
安装
npx skills add n8n-io/n8n --skill n8n-create-instance-ai-evalnpx skills add n8n-io/n8n --skill n8n-create-instance-ai-eval -a claude-code -gnpx skills add n8n-io/n8n --skill n8n-create-instance-ai-eval -a codex -ggemini skills install https://github.com/n8n-io/n8n.git --path .agents/skills/create-instance-ai-evalgh skill install n8n-io/n8n n8n-create-instance-ai-evalgit clone --depth 1 https://github.com/n8n-io/n8n /tmp/n8n-io__n8n
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/n8n-io__n8n/.agents/skills/create-instance-ai-eval ~/.claude/skills/n8n-create-instance-ai-eval # Claude Code
cp -R /tmp/n8n-io__n8n/.agents/skills/create-instance-ai-eval ~/.agents/skills/n8n-create-instance-ai-eval # Codex, Cursor, Gemini CLI, Copilot, OpenCode