{
  "metadata": {
    "skill_name": "smart-canteen-sku-standard-package",
    "skill_path": "/Users/jack/code/010-cpt/008-zhct/zhctprompt/standards-stack/agent-skills/skills/smart-canteen-sku-standard-package",
    "timestamp": "2026-09-01T12:47:49Z",
    "evals_run": [
      0
    ],
    "runs_per_configuration": 1
  },
  "runs": [
    {
      "eval_id": 0,
      "eval_name": "四个SKU必须形成双XMind加信息图",
      "configuration": "with_skill",
      "run_number": 1,
      "result": {
        "pass_rate": 1,
        "passed": 2,
        "failed": 0,
        "total": 2,
        "time_seconds": 56.366,
        "tokens": 165298,
        "errors": 0
      },
      "expectations": [
        {
          "text": "expect.must_contain",
          "passed": true,
          "evidence": "all checks passed"
        },
        {
          "text": "output_contains{all:[两个 XMind 信息图 NotebookLM 历史]}",
          "passed": true,
          "evidence": "output satisfies all contains checks (all:[两个 XMind 信息图 NotebookLM 历史])"
        }
      ]
    },
    {
      "eval_id": 0,
      "eval_name": "销售目录不得包含凭据和过程资料",
      "configuration": "with_skill",
      "run_number": 1,
      "result": {
        "pass_rate": 0,
        "passed": 0,
        "failed": 1,
        "total": 1,
        "time_seconds": 26.725,
        "tokens": 114093,
        "errors": 0
      },
      "expectations": [
        {
          "text": "expect.must_contain",
          "passed": false,
          "evidence": "output does not contain \"账号密码\""
        }
      ]
    },
    {
      "eval_id": 0,
      "eval_name": "项目现场一张图应路由到案例孪生",
      "configuration": "with_skill",
      "run_number": 1,
      "result": {
        "pass_rate": 1,
        "passed": 2,
        "failed": 0,
        "total": 2,
        "time_seconds": 51.771,
        "tokens": 270478,
        "errors": 0
      },
      "expectations": [
        {
          "text": "expect.must_contain",
          "passed": true,
          "evidence": "all checks passed"
        },
        {
          "text": "output_contains{all:[smart-canteen-evidence-backed-case-twin 现场 BOM]}",
          "passed": true,
          "evidence": "output satisfies all contains checks (all:[smart-canteen-evidence-backed-case-twin 现场 BOM])"
        }
      ]
    }
  ],
  "run_summary": {
    "with_skill": {
      "pass_rate": {
        "mean": 0.6666666666666666,
        "stddev": 0.4714045207910317,
        "min": 0,
        "max": 1
      },
      "time_seconds": {
        "mean": 44.95400000000001,
        "stddev": 13.025637207701841,
        "min": 26.725,
        "max": 56.366
      },
      "tokens": {
        "mean": 183289.66666666666,
        "stddev": 65099.1147319907,
        "min": 114093,
        "max": 270478
      }
    },
    "without_skill": null,
    "delta": null
  }
}