File size: 6,820 Bytes
4b045d4
 
 
 
 
 
 
 
 
b34e8a9
4b045d4
 
 
 
 
b34e8a9
4b045d4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b34e8a9
4b045d4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b34e8a9
 
 
 
 
 
 
 
 
 
 
 
 
4b045d4
 
b34e8a9
4b045d4
b34e8a9
4b045d4
 
 
b34e8a9
4b045d4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b34e8a9
 
 
 
 
 
 
 
 
 
 
 
 
 
4b045d4
 
 
 
 
 
 
 
 
b34e8a9
 
4b045d4
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
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
#!/usr/bin/env python3
"""
config-generator.py — 环境变量驱动的 openclaw.json 生成器

三层配置合并:
  1. 基础模板 (安全默认值)
  2. 从备份恢复的 openclaw.json
  3. 环境变量覆盖 (最高优先级)

端口: 18889 (内部端口,由 Caddy 反向代理)
"""

import json, os, copy

STATE_DIR = os.environ.get("OPENCLAW_STATE_DIR", "/root/.openclaw")
GATEWAY_PORT = int(os.environ.get("OPENCLAW_GATEWAY_PORT", 18889))

def deep_merge(base: dict, overlay: dict) -> dict:
    result = copy.deepcopy(base)
    for k, v in overlay.items():
        if k in result and isinstance(result[k], dict) and isinstance(v, dict):
            result[k] = deep_merge(result[k], v)
        else:
            result[k] = copy.deepcopy(v)
    return result

def _get_gw_password() -> str:
    pw = os.environ.get("OPENCLAW_GATEWAY_PASSWORD", "") or os.environ.get("OPENCLAW_GATEWAY_TOKEN", "")
    if not pw:
        import secrets
        pw = "openclaw-" + secrets.token_hex(8)
        print(f"[config] Generated gateway password: {pw}")
    return pw

def generate() -> dict:
    # ── Layer 1: 基础模板 ──
    config = {
        "models": {"providers": {}},
        "agents": {"defaults": {}, "list": []},
        "skills": {"entries": {}, "allowBundled": []},
        "plugins": {"entries": {}, "allow": []},
        "gateway": {
            "mode": "local",
            "bind": "lan",
            "port": GATEWAY_PORT,
            "trustedProxies": ["0.0.0.0/0"],
            "auth": {
                "mode": "token",
                "token": _get_gw_password(),
            },
            "controlUi": {
                "enabled": True,
                "allowInsecureAuth": True,
                "dangerouslyDisableDeviceAuth": True,
                "dangerouslyAllowHostHeaderOriginFallback": True,
            },
        },
        "browser": {
            "enabled": False,
            "headless": True,
        },
    }

    # ── Layer 2: 合并恢复的配置 ──
    restored_path = os.path.join(STATE_DIR, "openclaw.json")
    if os.path.exists(restored_path):
        try:
            with open(restored_path) as f:
                restored = json.load(f)
            config = deep_merge(config, restored)
        except Exception as e:
            print(f"[config] Warning: could not load restored config: {e}")

    # ── Layer 3: 环境变量覆盖 ──

    # 3a. 模型提供商 (只在有 API base + key 时生成)
    provider = os.environ.get("LLM_PROVIDER", "default")
    base_url = os.environ.get("OPENAI_API_BASE", "").rstrip("/")
    base_url = base_url.replace("/chat/completions", "")

    if base_url and os.environ.get("OPENAI_API_KEY"):
        config["models"]["providers"][provider] = {
            "baseUrl": base_url,
            "apiKey": os.environ.get("OPENAI_API_KEY", ""),
            "api": "openai-completions",
            "models": [
                {
                    "id": os.environ.get("MODEL", "gpt-4o"),
                    "name": os.environ.get("MODEL", "gpt-4o"),
                    "contextWindow": 128000,
                }
            ],
        }
        # 额外模型 (JSON array)
        extra = os.environ.get("EXTRA_MODELS", "")
        if extra:
            try:
                config["models"]["providers"][provider]["models"].extend(json.loads(extra))
            except Exception:
                pass
    else:
        # 无 API 配置时占位,避免配置验证失败
        config["models"]["providers"]["placeholder"] = {
            "baseUrl": "http://localhost:8080/v1",
            "apiKey": "not-configured-yet",
            "api": "openai-completions",
            "models": [{"id": "placeholder", "name": "placeholder", "contextWindow": 4096}],
        }
        print("[config] Warning: OPENAI_API_BASE or OPENAI_API_KEY not set, using placeholder")

    # 3b. 兼容旧版 plugins.allow → plugins.bundledDiscovery
    config["plugins"]["bundledDiscovery"] = "compat"

    # 3b. Agent 默认配置
    default_model = f"{provider}/{os.environ.get('MODEL', 'gpt-4o')}"
    config["agents"]["defaults"] = {
        "model": {"primary": default_model},
        "workspace": os.path.join(STATE_DIR, "workspace"),
    }

    # 3c. Agents: 5 role-based + 1 coordinator
    default_names = ["开发组-产品经理", "开发组-架构师", "开发组-开发", "开发组-测试", "开发组-项目经理", "总协调"]
    for i in range(1, 7):  # 1-5 role agents, 6 coordinator
        if i == 6:
            agent_id = "coordinator"
            name = os.environ.get("COORDINATOR_NAME", default_names[5])
            workspace = os.path.join(STATE_DIR, "workspace-coordinator")
            model_override = default_model
        else:
            agent_id = f"agent-{i}"
            name = os.environ.get(f"AGENT_{i}_NAME", default_names[i - 1])
            workspace = os.path.join(STATE_DIR, f"workspace-agent-{i}")
            model_override = os.environ.get(f"AGENT_{i}_MODEL", default_model)

        agent_def = {
            "id": agent_id,
            "name": name,
            "workspace": workspace,
            "model": {"primary": model_override},
        }

        # Skills
        skills_str = os.environ.get(f"AGENT_{i}_SKILLS", "")
        if skills_str:
            agent_def["skills"] = [s.strip() for s in skills_str.split(",")]

        config["agents"]["list"].append(agent_def)

    # 3d. Skills
    skills_enabled = os.environ.get("SKILLS_ENABLED", "")
    if skills_enabled:
        for s in skills_enabled.split(","):
            s = s.strip()
            if s:
                config["skills"]["entries"][s] = {"enabled": True}

    # 3e. Plugins
    plugins_allow = os.environ.get("PLUGINS_ALLOW", "")
    if plugins_allow:
        config["plugins"]["allow"] = [p.strip() for p in plugins_allow.split(",")]
    else:
        # Default plugin set (only bundled skills, not external plugins)
        config["plugins"]["allow"] = [
            "multi-search-cn",
            "github",
            "summarize",
        ]

    # Enable bundled skills (these are in plugins.entries as skills)
    core_skills = ["multi-search-cn", "github", "summarize"]
    for p in core_skills:
        if p not in config["plugins"]["allow"]:
            config["plugins"]["allow"].append(p)
        config["plugins"]["entries"][p] = {"enabled": True}

    return config

if __name__ == "__main__":
    config = generate()
    os.makedirs(STATE_DIR, exist_ok=True)
    out = os.path.join(STATE_DIR, "openclaw.json")
    with open(out, "w") as f:
        json.dump(config, f, indent=2, ensure_ascii=False)
    # Debug: output full config to logs
    print("[config] Full configuration (JSON):\n" + json.dumps(config, indent=2, ensure_ascii=False))
    print(f"[config] Generated → {out}")