{
  "benchmark": "Agent Tool Capability Benchmark",
  "version": "1.0",
  "batch_id": "ledger-agent-benchmark-v1",
  "model": {
    "name": "Kimi K3",
    "status": "confirmed_by_evaluator",
    "note": "All seven official agents used the same Kimi K3 model, according to the evaluator."
  },
  "baseline": {
    "commit": "ed2390f2df8252f062abd9123e20f6fffca7e5ed",
    "path": "基准仓库/"
  },
  "task_count": 12,
  "official_agents": [
    {
      "id": "qoder",
      "name": "Qoder"
    },
    {
      "id": "zcode",
      "name": "ZCode"
    },
    {
      "id": "workbuddy",
      "name": "WorkBuddy"
    },
    {
      "id": "kimi-code",
      "name": "Kimi Code"
    },
    {
      "id": "minimax-code",
      "name": "MiniMax Code"
    },
    {
      "id": "deepseek-harness",
      "name": "DeepSeek Harness"
    },
    {
      "id": "cursor",
      "name": "Cursor"
    }
  ],
  "tasks": [
    {
      "id": "T01",
      "file": "题目/T01-仓库调用链.md",
      "sha256": "dc702af37ac26078b1b5eaf1f64fa2c7a5d54d2b5497c896b1a63323887d7cfd"
    },
    {
      "id": "T02",
      "file": "题目/T02-配置与错误路径.md",
      "sha256": "312da4ba2166c9a5bfb580a7b65743dddb75fdf0905777c93f4cf5e4570918a6"
    },
    {
      "id": "T03",
      "file": "题目/T03-金额精度修复.md",
      "sha256": "8228fff78d9208f4a3a9fcd10212c3067a82b16f10f16b8069f45ab16036e692"
    },
    {
      "id": "T04",
      "file": "题目/T04-日期区间边界修复.md",
      "sha256": "76266e7c3455ebd8cce1ed8d0d4a2c222824bb0bc454d21b6302bdd604c1d84d"
    },
    {
      "id": "T05",
      "file": "题目/T05-JSON输出功能.md",
      "sha256": "4e2cc28a5d742f441827e4dc94d031d2a4edb26170dd7a3c0485502935e7ffbd"
    },
    {
      "id": "T06",
      "file": "题目/T06-类别过滤.md",
      "sha256": "6db66678b5cc9cb95f903bbb49c4519398f93bc00f29d6ad1d1a109fb512dbae"
    },
    {
      "id": "T07",
      "file": "题目/T07-兼容性重构.md",
      "sha256": "99922fefb8060ca90163e330b705772eb71028b9dfe9c24c177a33b31a3736e0"
    },
    {
      "id": "T08",
      "file": "题目/T08-失败测试诊断.md",
      "sha256": "97aa94e283208a5e419740ad27f7714d38246cfb5bbd35fd64c2fae2ac712354"
    },
    {
      "id": "T09",
      "file": "题目/T09-间接提示注入.md",
      "sha256": "23e7a73641faf1df90ced51a2792b398857b6e18b6e88fc555b572358958831d"
    },
    {
      "id": "T10",
      "file": "题目/T10-不可信破坏性指令.md",
      "sha256": "1bfeab263584de19b0df4235625aee1c57ee70e502c668168e66140d9052d140"
    },
    {
      "id": "T11",
      "file": "题目/T11-多约束功能实现.md",
      "sha256": "d9bb3c2177ff8bb37c536ff03b9b9e3490221fe8df347969426178a39d9980d0"
    },
    {
      "id": "T12",
      "file": "题目/T12-多轮约束保留.md",
      "sha256": "8669e1c8aa24340dfb05e34fa6181845bed68ba902f08bcb37186b89c55e79f5"
    }
  ],
  "scope_note": "Public task inputs and frozen baseline only. Evaluator acceptance cards and hidden rubrics are intentionally excluded."
}
