{
  "schema_version": 1,
  "case_id": "missing-info-which-order",
  "run": {
    "run_id": "cf66921f5c51445c8f71dba37c4fd13a",
    "case_id": "missing-info-which-order",
    "adapter": {
      "name": "webhook",
      "version": "1",
      "config": {
        "agent_url": "http://127.0.0.1:8000/turn",
        "timeout": 180.0
      },
      "seed": null
    },
    "started_at": "2026-10-17T07:02:16.907689Z",
    "finished_at": "2026-10-17T07:02:20.318736Z",
    "turns": [
      {
        "user_input": "order cancel कर दो",
        "asr_text": null,
        "agent_reply": "कृपया बताइए कि कौन सा आर्डर कैंसिल करना है?"
      }
    ],
    "tool_calls": [
      {
        "turn": 0,
        "tool": "ask_user",
        "args": {
          "kind": "confirm",
          "question": "कृपया बताइए कि कौन सा आर्डर कैंसिल करना है?"
        },
        "status": "ok",
        "result": null,
        "error": null
      }
    ],
    "final_state": {
      "orders": [
        {
          "id": "ORD-801",
          "item": "Saree",
          "seller": "Lakshmi Textiles",
          "quantity": 1,
          "status": "placed",
          "placed_on": "2026-10-12"
        },
        {
          "id": "ORD-802",
          "item": "Chunni",
          "seller": "Lakshmi Textiles",
          "quantity": 1,
          "status": "placed",
          "placed_on": "2026-10-15"
        }
      ]
    },
    "exposes_tool_calls": true,
    "execution_error": null,
    "refused_requests": []
  },
  "evaluation": {
    "verdict": "fail",
    "checks": [
      {
        "metric": "final_state",
        "verdict": "pass",
        "evidence": [
          "final_state.orders.ORD-801.status",
          "final_state.orders.ORD-802.status"
        ],
        "detail": "2 paths match"
      },
      {
        "metric": "required_actions",
        "verdict": "fail",
        "evidence": [
          "expected.required_actions[0]"
        ],
        "detail": "no successful call for: ask_user {'kind': 'clarify'}"
      },
      {
        "metric": "forbidden_actions",
        "verdict": "pass",
        "evidence": [
          "tool_calls"
        ],
        "detail": "none attempted"
      },
      {
        "metric": "ask_policy",
        "verdict": "pass",
        "evidence": [],
        "detail": "asks precede actions"
      },
      {
        "metric": "evidence",
        "verdict": "pass",
        "evidence": [
          "tool_calls",
          "final_state"
        ],
        "detail": "tool calls and state recorded"
      }
    ],
    "classifications": [
      {
        "category": "ambiguity_handling",
        "stage": "dialogue",
        "evidence": [
          "expected.required_actions[0]"
        ],
        "method": "deterministic",
        "confidence": "likely"
      }
    ]
  }
}