✔ before: POST http://localhost:8800/cache-clear · HTTP 200 · 0.0s · {"cleared": 2}
✔ OpenAI credential: broken key (drill). Restore with: uv run tools/flow.py llm up
POST http://localhost:5678/webhook/my-helpdesk · 4 case(s) × 1
✘ VPN · FAQ 원문으로 · 0.9s · status 500 (expected 200); route = null (expected 간단 답변); answer has none of ['인증서']
    → request  POST /webhook/my-helpdesk
      {
        "question": "VPN 접속이 자꾸 끊겨요. 어떻게 하나요?",
        "user": "이준호",
        "no_cache": true
      }
    ← response HTTP 500
      {
        "message": "Error in workflow"
      }
✘ 접수 현황 · 상태 그대로 · 0.4s · status 500 (expected 200); route = null (expected 간단 답변); answer has none of ['처리 중']
    → request  POST /webhook/my-helpdesk
      {
        "question": "T-1002 처리 상태 알려 주세요",
        "user": "박서연"
      }
    ← response HTTP 500
      {
        "message": "Error in workflow"
      }
✘ 모르는 질문 · 담당자 · 0.9s · status 500 (expected 200); route = null (expected 담당자 연결)
    → request  POST /webhook/my-helpdesk
      {
        "question": "오늘 점심 메뉴 추천해 주세요",
        "user": "이준호"
      }
    ← response HTTP 500
      {
        "message": "Error in workflow"
      }
✔ 빈 질문 · 0.0s
    → request  POST /webhook/my-helpdesk
      {
        "question": "",
        "user": "이준호"
      }
    ← response HTTP 400
      {
        "error": "question is required"
      }
Passed 1/4
report: runs/lab03-llm-down-20261024-175047.json
✔ OpenAI credential: real key from .env restored