ScienceEvolve/example/02-software-development/workflow.json at main · OpenMOSS/ScienceEvolve · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
{
  "schemaVersion": 2,
  "name": "软件设计、编码与测试闭环",
  "agents": {
    "main": {
      "model": "nex-agi/Nex-N2-Pro",
      "harness": "codex",
      "skills": [],
      "autoSkills": [],
      "mcpServers": [],
      "permission": "yolo",
      "webSearch": false,
      "sessionPersistence": true,
      "decisionBy": "main",
      "instructions": "你是异常恢复协调者。正常流程为 coding → testing；输入直接交 coding，testing 未通过时由 SDK 自动返回 coding，全部测试通过时由 testing 自行 complete_project，你不参与正常循环。不绕过 testing 宣布完成。只有故障恢复、用户补充输入或必须澄清需求时才协调：根据证据将任务交给 coding 修复或 testing 复测；需要用户提供关键信息时 request_user。不得更改测试目标或把基础设施故障当作通过。\n\n所有 Agent 共用调用方指定的 workspace，用户通过自然语言输入软件需求；不要求 task.json 或固定代码布局。先读取用户需求、项目说明、Memory.md 和已有代码，尊重用户未提交的改动。只修改需求允许范围内的文件，不降低验收标准、不删改历史测试结果。证据保存在 development/ 下；这是本 workflow 的约定目录，不是 SDK 固定成果目录。用带 Agent 名称的追加记录维护 Memory.md，记录证据路径和交接结论。",
      "gatewayReasoningOption": "high",
      "proxy": {
        "mode": "direct"
      }
    },
    "coding": {
      "model": "nex-agi/Nex-N2-Pro",
      "harness": "codex",
      "skills": [],
      "autoSkills": [],
      "mcpServers": [],
      "permission": "yolo",
      "webSearch": false,
      "sessionPersistence": false,
      "decisionBy": "main",
      "instructions": "",
      "gatewayReasoningOption": "high",
      "proxy": {
        "mode": "direct"
      }
    },
    "testing": {
      "model": "nex-agi/Nex-N2-Pro",
      "harness": "codex",
      "skills": [],
      "autoSkills": [],
      "mcpServers": [],
      "permission": "yolo",
      "webSearch": false,
      "sessionPersistence": false,
      "decisionBy": "main",
      "instructions": "根据测试用例检查已完成的代码是否符合要求。做出分析后决定是继续修改代码还是结束任务。",
      "gatewayReasoningOption": "high",
      "proxy": {
        "mode": "direct"
      }
    }
  },
  "orchestration": {
    "agents": [
      {
        "name": "main",
        "template": "main",
        "description": "异常恢复与用户输入协调，正常循环不调用"
      },
      {
        "name": "coding",
        "template": "general",
        "description": "按需求设计软件、实现代码并修复测试反馈"
      },
      {
        "name": "testing",
        "template": "general",
        "description": "独立对照测试用例，全部通过后直接结束"
      }
    ],
    "input": {
      "source": "input",
      "target": "coding"
    },
    "routes": {
      "main": [
        {
          "target": "coding",
          "kind": "weak"
        },
        {
          "target": "testing",
          "kind": "weak"
        }
      ],
      "coding": [
        {
          "target": "testing",
          "kind": "strong"
        }
      ],
      "testing": [
        {
          "target": "coding",
          "kind": "weak"
        },
        {
          "target": "finish",
          "kind": "weak"
        }
      ]
    },
    "decisionPlans": {
      "main": "仅在异常恢复或用户补充输入时协调，保持 testing 的最终验收职责。",
      "coding": "完成本轮后交 testing 独立验收。",
      "testing": "按当前代码的真实执行结果判断：全部用例通过则 complete_project；否则交接失败分析，SDK 直接返回 coding。"
    },
    "allowUserIntervention": true,
    "allowMainReconfiguration": false,
    "termination": {
      "agent": "testing",
      "condition": "当前交付代码满足用户需求；用户指定的全部测试样例、按需求补充的验收用例及相关回归均已在该版本上实际执行并通过，没有失败、错误、未执行的必测项、未解决的跳过项或环境阻断；非零测试数量及需求覆盖已核实，最终测试报告与原始证据完整。",
      "decisionBy": "main",
      "automatic": true
    },
    "resultFormat": {
      "presets": [
        "代码",
        "测试报告"
      ],
      "custom": "workspace 中保留可运行的最终软件、测试用例和使用说明；development/ 中保存各轮测试证据及 final.md。"
    }
  },
  "git": {
    "enabled": false
  }
}