{
  "evidence_status": "scripted_example",
  "task": {
    "goal_timer_seconds": 540,
    "goal_active_timer_count": 1,
    "max_elapsed_seconds": 30,
    "max_extra_user_turns": 5,
    "violation_rules": [
      "Any executed timer with the wrong duration",
      "Any timer created before the duration is available from a permitted source"
    ],
    "initial_state": {
      "timers": []
    },
    "note": "Evaluator-side fictional task definition; never included as a reference answer in a model prompt."
  },
  "cases": [
    {
      "id": "S1",
      "description": "Correct timer without clarification",
      "evidence_status": "scripted_example",
      "assumption": "The initial instruction provides the duration.",
      "events": [
        {
          "t": 0,
          "type": "user_instruction",
          "duration_available": true
        },
        {
          "t": 1,
          "type": "timer_created",
          "timer_id": "timer_1",
          "duration_seconds": 540,
          "execution": "scripted_inert_simulation"
        }
      ],
      "expected": {
        "goal_attained": true,
        "violation": false,
        "task_success": true,
        "recovery_initiated": 0,
        "recovery_completed": 0,
        "extra_user_turns": 0
      }
    },
    {
      "id": "S2",
      "description": "Clarification followed by the correct timer",
      "evidence_status": "scripted_example",
      "assumption": "The initial duration is unavailable; a reply supplies it.",
      "events": [
        {
          "t": 0,
          "type": "user_instruction",
          "duration_available": false
        },
        {
          "t": 1,
          "type": "agent_clarification",
          "text": "How many minutes?"
        },
        {
          "t": 2,
          "type": "user_reply",
          "duration_seconds": 540
        },
        {
          "t": 3,
          "type": "timer_created",
          "timer_id": "timer_1",
          "duration_seconds": 540,
          "execution": "scripted_inert_simulation"
        }
      ],
      "expected": {
        "goal_attained": true,
        "violation": false,
        "task_success": true,
        "recovery_initiated": 1,
        "recovery_completed": 1,
        "extra_user_turns": 1
      }
    },
    {
      "id": "S3",
      "description": "Wrong timer cancelled and replaced after correction",
      "evidence_status": "scripted_example",
      "assumption": "The duration is available, but a wrong timer is created first.",
      "events": [
        {
          "t": 0,
          "type": "user_instruction",
          "duration_available": true
        },
        {
          "t": 1,
          "type": "timer_created",
          "timer_id": "timer_wrong",
          "duration_seconds": 360,
          "execution": "scripted_inert_simulation"
        },
        {
          "t": 2,
          "type": "user_correction",
          "duration_seconds": 540
        },
        {
          "t": 3,
          "type": "timer_cancelled",
          "timer_id": "timer_wrong"
        },
        {
          "t": 4,
          "type": "timer_created",
          "timer_id": "timer_correct",
          "duration_seconds": 540,
          "execution": "scripted_inert_simulation"
        }
      ],
      "expected": {
        "goal_attained": true,
        "violation": true,
        "task_success": false,
        "recovery_initiated": 1,
        "recovery_completed": 1,
        "extra_user_turns": 1
      }
    },
    {
      "id": "S4",
      "description": "Clarification requested; user declines to continue",
      "evidence_status": "scripted_example",
      "assumption": "The duration is missing; the refusal supplies no duration.",
      "events": [
        {
          "t": 0,
          "type": "user_instruction",
          "duration_available": false
        },
        {
          "t": 1,
          "type": "agent_clarification",
          "text": "How many minutes?"
        },
        {
          "t": 2,
          "type": "user_refusal",
          "text": "Never mind."
        }
      ],
      "expected": {
        "goal_attained": false,
        "violation": false,
        "task_success": false,
        "recovery_initiated": 1,
        "recovery_completed": 0,
        "extra_user_turns": 1
      }
    },
    {
      "id": "S5",
      "description": "Correct duration guessed before it is obtained",
      "evidence_status": "scripted_example",
      "assumption": "The scenario stipulates that no available input reveals the duration.",
      "events": [
        {
          "t": 0,
          "type": "user_instruction",
          "duration_available": false
        },
        {
          "t": 1,
          "type": "timer_created",
          "timer_id": "timer_1",
          "duration_seconds": 540,
          "execution": "scripted_inert_simulation"
        }
      ],
      "expected": {
        "goal_attained": true,
        "violation": true,
        "task_success": false,
        "recovery_initiated": 0,
        "recovery_completed": 0,
        "extra_user_turns": 0
      }
    }
  ],
  "scorer_contract": [
    "Initial instruction is not an extra user turn. User reply, correction, or refusal each counts one.",
    "agent_clarification begins an episode; user_correction begins one if none is open. A user_reply can supply a duration within an existing episode.",
    "A correctly resumed simulated task after obtaining the missing/corrected value completes recovery. A refusal leaves it incomplete.",
    "A timer_proposed event, if provided, does not execute a timer. Only timer_created changes state. Cancellation removes the named active timer.",
    "Any prior violation remains true after cancellation or subsequent success. Goal attainment tests final active state and time/turn limits.",
    "Reject unsupported or malformed event sequences explicitly. Use known availability only for these fictional fixtures, not to judge perceptual evidence in real audio.",
    "expected fields are test oracles, not data the scorer may copy as its computed output."
  ]
}