{
  "schema": "adoptlab-static-evidence-v1",
  "recorded_date": "2026-09-30",
  "model": "deepseek-flash",
  "materials": {
    "A": {
      "guide": "Use the four record tools to finish the supplied task and save outputs.",
      "descriptions": {
        "list_fixture_files": "List input files.",
        "read_records": "Read records.",
        "normalize_filter_records": "Normalize and filter records.",
        "write_outputs": "Write result and summary."
      }
    },
    "B": {
      "guide": "1. List fixtures and read records.json. 2. Pass its records array and the task's min_cents (integer cents) to normalize_filter_records. 3. Pass the returned result and summary to write_outputs. Validate every row before filtering. Treat tool errors as actionable input errors; report their error code without fabricating output. CNY/USD totals remain separate. Dates require a timezone and normalize to UTC. All paths are relative.",
      "descriptions": {
        "list_fixture_files": "List available synthetic input files in fixtures; no host files are exposed.",
        "read_records": "Read JSON input using a relative path, e.g. records.json; returns a records array.",
        "normalize_filter_records": "Validate ALL records, normalize whitespace/case/money/timezones, then filter active rows >= min_cents (integer cents). Returns result and summary for write_outputs. Invalid input returns a stable error code.",
        "write_outputs": "Write BOTH result.json and summary.json into this run's outputs directory. Pass the result and summary from normalize_filter_records; return output hashes."
      }
    }
  },
  "cases": [
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.014382
    },
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.016058
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01383
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "A",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012574
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "A",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013126
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.015246
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014678
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "A",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01282
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "A",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013036
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014922
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.016234
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "A",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014482
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.011772
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012602
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014686
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.013292
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.010934
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01249
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01214
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.011104
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.013192
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.010156
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "B",
      "trial": 1,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.009554
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "A",
      "trial": 1,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.011618
    },
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.015528
    },
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014136
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012574
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013942
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.015278
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013166
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012628
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014268
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014448
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013036
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014186
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.016428
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012584
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01139
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "A",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01305
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01496
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012442
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "A",
      "trial": 2,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.010724
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "A",
      "trial": 2,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.010612
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012344
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.009386
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "A",
      "trial": 2,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.013086
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "A",
      "trial": 2,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.011872
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "B",
      "trial": 2,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.04887
    },
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014204
    },
    {
      "task_id": "task-01",
      "family": "threshold",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.015312
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014132
    },
    {
      "task_id": "task-02",
      "family": "threshold",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012574
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.013142
    },
    {
      "task_id": "task-03",
      "family": "whitespace",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.015278
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01439
    },
    {
      "task_id": "task-04",
      "family": "whitespace",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01282
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012876
    },
    {
      "task_id": "task-05",
      "family": "currency",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014848
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.017012
    },
    {
      "task_id": "task-06",
      "family": "currency",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014186
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.011422
    },
    {
      "task_id": "task-07",
      "family": "timezone",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012584
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.014672
    },
    {
      "task_id": "task-08",
      "family": "timezone",
      "material": "A",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.01341
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "A",
      "trial": 3,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.010718
    },
    {
      "task_id": "task-09",
      "family": "empty",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012498
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "artifact",
      "reason": "matched",
      "cost_upper_cny": 0.012344
    },
    {
      "task_id": "task-10",
      "family": "empty",
      "material": "A",
      "trial": 3,
      "passed": false,
      "kind": "artifact",
      "reason": "output_mismatch",
      "cost_upper_cny": 0.010822
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "A",
      "trial": 3,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.013198
    },
    {
      "task_id": "task-11",
      "family": "invalid",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.009386
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "B",
      "trial": 3,
      "passed": true,
      "kind": "correct_rejection",
      "reason": "expected_error_observed",
      "cost_upper_cny": 0.009182
    },
    {
      "task_id": "task-12",
      "family": "invalid",
      "material": "A",
      "trial": 3,
      "passed": false,
      "kind": "correct_rejection",
      "reason": "rejection_mismatch",
      "cost_upper_cny": 0.011774
    }
  ],
  "limitations": "One model, synthetic records, correlated repeats and bundled guide/description changes. This is a recorded explorer, with no live inference or unique-user claim."
}