openclaw-agent-dashboard 1.0.21 → 1.0.23

package/scripts/test_available_models.py

@@ -1,284 +0,0 @@
-#!/usr/bin/env python3
-"""
-测试模型配置：get_all_available_models、update_agent_model 与 OpenClaw 配置格式一致性。
-用法: cd 项目根目录 && python3 scripts/test_available_models.py
-"""
-import json
-import os
-import sys
-import tempfile
-from pathlib import Path
-
-# 添加 backend 到 path
-ROOT = Path(__file__).resolve().parent.parent
-sys.path.insert(0, str(ROOT / "src" / "backend"))
-
-
-def run_test_env(config: dict):
-    """创建临时 openclaw.json 并设置 OPENCLAW_HOME，返回 (tmp_dir, old_home)"""
-    tmp_dir = tempfile.mkdtemp()
-    config_path = Path(tmp_dir) / "openclaw.json"
-    with open(config_path, 'w', encoding='utf-8') as f:
-        json.dump(config, f, indent=2, ensure_ascii=False)
-    old_home = os.environ.get('OPENCLAW_HOME')
-    os.environ['OPENCLAW_HOME'] = tmp_dir
-    for mod in list(sys.modules.keys()):
-        if 'data.config_reader' in mod or 'data.agent_config_manager' in mod:
-            del sys.modules[mod]
-    return tmp_dir, old_home, config_path
-
-
-def run_test(name: str, config: dict) -> list:
-    """用临时 openclaw.json 运行 get_all_available_models"""
-    tmp_dir, old_home, _ = run_test_env(config)
-    try:
-        from data.agent_config_manager import get_all_available_models
-        return get_all_available_models()
-    finally:
-        if old_home is not None:
-            os.environ['OPENCLAW_HOME'] = old_home
-        else:
-            os.environ.pop('OPENCLAW_HOME', None)
-        import shutil
-        shutil.rmtree(tmp_dir, ignore_errors=True)
-
-
-def main():
-    print("=" * 60)
-    print("测试：模型列表（白名单 / providers / agents 收集）")
-    print("=" * 60)
-
-    # 场景1: agents.defaults.model + agents.list 中 agent 有 model
-    config1 = {
-        "agents": {
-            "defaults": {
-                "model": {
-                    "primary": "zhipu/glm-4",
-                    "fallbacks": ["zhipu/glm-4-flash"]
-                }
-            },
-            "list": [
-                {"id": "main", "name": "PM", "model": {"primary": "zhipu/glm-4.5", "fallbacks": []}},
-                {"id": "analyst", "name": "BA"}
-            ]
-        }
-    }
-    # 注意：没有 models.providers
-
-    print("\n场景1: defaults.model + list 中部分 agent 有 model")
-    print("  defaults: primary=zhipu/glm-4, fallbacks=[zhipu/glm-4-flash]")
-    print("  main: primary=zhipu/glm-4.5")
-    print("  analyst: 无 model，继承 defaults")
-    try:
-        r1 = run_test("scene1", config1)
-        ids = [m['id'] for m in r1]
-        print(f"  结果: {ids}")
-        expected = {"zhipu/glm-4.5", "zhipu/glm-4", "zhipu/glm-4-flash"}
-        if set(ids) == expected:
-            print("  ✓ 通过")
-        else:
-            print(f"  期望: {expected}, 实际: {set(ids)}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-        import traceback
-        traceback.print_exc()
-
-    # 场景2: 仅 defaults，list 中 agent 都无 model
-    config2 = {
-        "agents": {
-            "defaults": {
-                "model": {"primary": "openai/gpt-4", "fallbacks": []}
-            },
-            "list": [
-                {"id": "main", "name": "PM"}
-            ]
-        }
-    }
-    print("\n场景2: 仅 defaults.model，agent 无 model")
-    try:
-        r2 = run_test("scene2", config2)
-        ids = [m['id'] for m in r2]
-        print(f"  结果: {ids}")
-        if ids == ["openai/gpt-4"]:
-            print("  ✓ 通过")
-        else:
-            print(f"  期望: ['openai/gpt-4'], 实际: {ids}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-
-    # 场景3: agents.list 为空
-    config3 = {
-        "agents": {
-            "defaults": {"model": {"primary": "zhipu/glm-4", "fallbacks": []}},
-            "list": []
-        }
-    }
-    print("\n场景3: agents.list 为空，仅有 defaults.model")
-    try:
-        r3 = run_test("scene3", config3)
-        ids = [m['id'] for m in r3]
-        print(f"  结果: {ids}")
-        if set(ids) == {"zhipu/glm-4"}:
-            print("  ✓ 通过")
-        else:
-            print(f"  期望: ['zhipu/glm-4'], 实际: {ids}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-
-    # 场景3b: 有 defaults.models 但 model 中未用
-    config3b = {
-        "agents": {
-            "defaults": {
-                "model": {"primary": "zhipu/glm-4", "fallbacks": []},
-                "models": {
-                    "zhipu/glm-4": {"alias": "GLM-4"},
-                    "zhipu/glm-4-flash": {"alias": "GLM-4 Flash"},
-                    "moonshot/kimi-k2": {"alias": "Kimi K2"}
-                }
-            },
-            "list": []
-        }
-    }
-    print("\n场景3b: agents.defaults.models 作为白名单（有则仅显示白名单）")
-    try:
-        r3b = run_test("scene3b", config3b)
-        ids = [m['id'] for m in r3b]
-        names = [m['name'] for m in r3b]
-        print(f"  结果: {ids}")
-        print(f"  展示名(name): {names} (应使用 id 不用别名)")
-        expected = {"zhipu/glm-4", "zhipu/glm-4-flash", "moonshot/kimi-k2"}
-        if set(ids) == expected:
-            # 展示名应为 id 简短形式，非 alias
-            expected_names = {"glm-4", "glm-4-flash", "kimi-k2"}
-            if set(names) == expected_names:
-                print("  ✓ 通过（白名单 + 展示用 id）")
-            else:
-                print(f"  ⚠ id 正确，但 name 应为 id 形式: {expected_names}, 实际: {set(names)}")
-        else:
-            print(f"  期望: {sorted(expected)}, 实际: {ids}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-
-    # 场景3c: 有 providers + 白名单，应仅显示白名单
-    config3c = {
-        "models": {
-            "providers": {
-                "zhipu": {"models": [{"id": "glm-4"}, {"id": "glm-4-flash"}, {"id": "glm-4.5"}]},
-                "openai": {"models": [{"id": "gpt-4"}]}
-            }
-        },
-        "agents": {
-            "defaults": {
-                "model": {"primary": "zhipu/glm-4", "fallbacks": []},
-                "models": {
-                    "zhipu/glm-4": {"alias": "GLM-4"},
-                    "zhipu/glm-4-flash": {"alias": "Flash"}
-                }
-            },
-            "list": [{"id": "main", "name": "PM"}]
-        }
-    }
-    print("\n场景3c: 有 providers + 白名单，应仅显示白名单中的模型")
-    try:
-        r3c = run_test("scene3c", config3c)
-        ids = [m['id'] for m in r3c]
-        print(f"  结果: {ids}")
-        expected = {"zhipu/glm-4", "zhipu/glm-4-flash"}
-        if set(ids) == expected:
-            print("  ✓ 通过")
-        else:
-            print(f"  期望: {sorted(expected)}, 实际: {ids}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-
-    # 场景3e: 协作流程模型面板应仅显示 agent 配置的（不含白名单未使用的）
-    config3e = {
-        "agents": {
-            "defaults": {
-                "model": {"primary": "zhipu/glm-4", "fallbacks": []},
-                "models": {
-                    "zhipu/glm-4": {"alias": "GLM-4"},
-                    "zhipu/glm-4-flash": {"alias": "Flash"},
-                    "moonshot/kimi-k2": {"alias": "Kimi"}
-                }
-            },
-            "list": [{"id": "main", "name": "PM"}]
-        }
-    }
-    print("\n场景3e: 协作流程模型面板（get_models_configured_by_agents）仅 agent 配置")
-    tmp_dir, old_home, _ = run_test_env(config3e)
-    try:
-        from data.config_reader import get_models_configured_by_agents
-        configured = get_models_configured_by_agents()
-        print(f"  结果: {configured}")
-        if configured == ["zhipu/glm-4"]:
-            print("  ✓ 通过（仅 primary，不含白名单中未用的 kimi-k2、glm-4-flash）")
-        else:
-            print(f"  期望: ['zhipu/glm-4'], 实际: {configured}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-    finally:
-        if old_home is not None:
-            os.environ['OPENCLAW_HOME'] = old_home
-        else:
-            os.environ.pop('OPENCLAW_HOME', None)
-        import shutil
-        shutil.rmtree(tmp_dir, ignore_errors=True)
-
-    # 场景3d: 配置保存格式（update_agent_model 写入格式与 OpenClaw 一致）
-    config3d = {
-        "agents": {
-            "defaults": {"model": {"primary": "zhipu/glm-4", "fallbacks": []}},
-            "list": [{"id": "main", "name": "PM", "model": {"primary": "zhipu/glm-4", "fallbacks": []}}]
-        }
-    }
-    print("\n场景3d: 配置保存格式（OpenClaw agents.list[].model.primary/fallbacks）")
-    tmp_dir, old_home, config_path = run_test_env(config3d)
-    try:
-        from data.agent_config_manager import update_agent_model, load_full_config
-        res = update_agent_model("main", primary="openai/gpt-4", fallbacks=["zhipu/glm-4-flash"])
-        if not res.get('success'):
-            print(f"  ✗ update_agent_model 失败: {res.get('error')}")
-        else:
-            saved = json.loads(config_path.read_text(encoding='utf-8'))
-            agent = next((a for a in saved.get('agents', {}).get('list', []) if a.get('id') == 'main'), None)
-            primary = agent and agent.get('model', {}).get('primary')
-            fallbacks = agent and agent.get('model', {}).get('fallbacks', [])
-            if primary == "openai/gpt-4" and fallbacks == ["zhipu/glm-4-flash"]:
-                print("  ✓ 通过（格式符合 OpenClaw agents.list[].model）")
-            else:
-                print(f"  期望 primary=openai/gpt-4 fallbacks=[zhipu/glm-4-flash]")
-                print(f"  实际 primary={primary} fallbacks={fallbacks}")
-    except Exception as e:
-        print(f"  ✗ 异常: {e}")
-    finally:
-        if old_home is not None:
-            os.environ['OPENCLAW_HOME'] = old_home
-        else:
-            os.environ.pop('OPENCLAW_HOME', None)
-        import shutil
-        shutil.rmtree(tmp_dir, ignore_errors=True)
-
-    # 场景4: 使用真实 openclaw.json（若存在）
-    real_path = Path.home() / ".openclaw" / "openclaw.json"
-    if real_path.exists():
-        print("\n场景4: 使用真实 ~/.openclaw/openclaw.json")
-        os.environ.pop('OPENCLAW_HOME', None)
-        try:
-            from data.agent_config_manager import get_all_available_models
-            r4 = get_all_available_models()
-            print(f"  结果: {len(r4)} 个模型")
-            for m in r4[:5]:
-                print(f"    - {m['id']} ({m['name']})")
-            if len(r4) > 5:
-                print(f"    ... 共 {len(r4)} 个")
-        except Exception as e:
-            print(f"  ✗ 异常: {e}")
-    else:
-        print("\n场景4: 跳过（无 ~/.openclaw/openclaw.json）")
-
-    print("\n" + "=" * 60)
-
-
-if __name__ == "__main__":
-    main()

package/scripts/test_version_display.sh

@@ -1,128 +0,0 @@
-#!/bin/bash
-# 版本号显示功能 - 快速测试脚本
-
-set -e
-
-PROJECT_DIR="/home/ubuntu/vrt-projects/projects/openclaw-agent-dashboard"
-cd "$PROJECT_DIR"
-
-echo "=================================="
-echo "版本号显示功能 - 快速测试"
-echo "=================================="
-echo ""
-
-# 颜色定义
-GREEN='\033[0;32m'
-RED='\033[0;31m'
-YELLOW='\033[1;33m'
-NC='\033[0m' # No Color
-
-# 测试函数
-test_passed() {
-    echo -e "${GREEN}✅ 通过${NC}: $1"
-}
-
-test_failed() {
-    echo -e "${RED}❌ 失败${NC}: $1"
-}
-
-test_info() {
-    echo -e "${YELLOW}ℹ️  信息${NC}: $1"
-}
-
-echo "1. 检查文件是否存在"
-echo "----------------------------------"
-
-FILES=(
-    "src/backend/data/version_info_reader.py"
-    "src/backend/api/version.py"
-    "frontend/src/components/common/VersionDisplay.vue"
-    ".staging/traceability_manifest.json"
-    ".staging/VERSION_DISPLAY_implementation_summary.md"
-)
-
-for file in "${FILES[@]}"; do
-    if [ -f "$file" ]; then
-        test_passed "文件存在: $file"
-    else
-        test_failed "文件缺失: $file"
-    fi
-done
-
-echo ""
-echo "2. 检查 Python 语法"
-echo "----------------------------------"
-
-python3 -m py_compile src/backend/data/version_info_reader.py && \
-    test_passed "version_info_reader.py 语法正确" || \
-    test_failed "version_info_reader.py 语法错误"
-
-python3 -m py_compile src/backend/api/version.py && \
-    test_passed "version.py 语法正确" || \
-    test_failed "version.py 语法错误"
-
-echo ""
-echo "3. 检查 API 路由注册"
-echo "----------------------------------"
-
-if grep -q "from api import.*version" src/backend/main.py; then
-    test_passed "main.py 已导入 version 模块"
-else
-    test_failed "main.py 未导入 version 模块"
-fi
-
-if grep -q "app.include_router(version.router" src/backend/main.py; then
-    test_passed "main.py 已注册 version 路由"
-else
-    test_failed "main.py 未注册 version 路由"
-fi
-
-echo ""
-echo "4. 检查前端组件集成"
-echo "----------------------------------"
-
-if grep -q "import VersionDisplay from './components/common/VersionDisplay.vue'" frontend/src/App.vue; then
-    test_passed "App.vue 已导入 VersionDisplay 组件"
-else
-    test_failed "App.vue 未导入 VersionDisplay 组件"
-fi
-
-if grep -q "<VersionDisplay />" frontend/src/App.vue; then
-    test_passed "App.vue 模板包含 VersionDisplay 标签"
-else
-    test_failed "App.vue 模板缺少 VersionDisplay 标签"
-fi
-
-echo ""
-echo "5. 检查版本号数据源"
-echo "----------------------------------"
-
-if [ -f "package.json" ]; then
-    VERSION=$(python3 -c "import json; print(json.load(open('package.json')).get('version', 'unknown'))")
-    test_info "当前版本号: $VERSION"
-    test_passed "package.json 存在且可解析"
-else
-    test_failed "package.json 不存在"
-fi
-
-echo ""
-echo "6. 检查组件目录结构"
-echo "----------------------------------"
-
-if [ -d "frontend/src/components/common" ]; then
-    test_passed "common 目录已创建"
-else
-    test_failed "common 目录缺失"
-fi
-
-echo ""
-echo "=================================="
-echo "测试完成"
-echo "=================================="
-echo ""
-echo "后续步骤："
-echo "1. 启动服务: npm start"
-echo "2. 测试 API: curl http://localhost:8000/api/version"
-echo "3. 打开浏览器访问: http://localhost:8000"
-echo "4. 检查右下角是否显示版本号"
-echo ""

package/scripts/test_websocket_ping.py

@@ -1,44 +0,0 @@
-#!/usr/bin/env python3
-"""测试 WebSocket JSON 心跳"""
-import asyncio
-import json
-import sys
-
-try:
-    import websockets
-except ImportError:
-    print("请安装: pip install websockets")
-    sys.exit(1)
-
-
-async def test():
-    uri = "ws://localhost:38272/ws"
-    print("连接 WebSocket...")
-    async with websockets.connect(uri, close_timeout=2) as ws:
-        # 1. 可能收到 full_state（若后端正常），或直接进入 ping 测试
-        # 先发送 JSON ping 测试心跳逻辑
-        print("1. 发送 JSON ping...")
-        await ws.send(json.dumps({"type": "ping", "timestamp": 1234567}))
-        msg = await asyncio.wait_for(ws.recv(), timeout=3)
-        data = json.loads(msg)
-        if data.get("type") != "pong":
-            print(f"FAIL: 期望 pong, 收到 {data.get('type')}")
-            return False
-        print("2. JSON ping -> pong: OK")
-
-        # 3. 发送纯文本 ping
-        await ws.send("ping")
-        msg = await asyncio.wait_for(ws.recv(), timeout=3)
-        data = json.loads(msg)
-        if data.get("type") != "pong":
-            print(f"FAIL: 期望 pong, 收到 {data.get('type')}")
-            return False
-        print("3. 文本 ping -> pong: OK")
-
-    print("所有 WebSocket 测试通过!")
-    return True
-
-
-if __name__ == "__main__":
-    ok = asyncio.run(test())
-    sys.exit(0 if ok else 1)

package/session_registry.json

@@ -1,58 +0,0 @@
-{
-  "schema_version": "1.0",
-  "project_id": "openclaw-agent-dashboard",
-  "updated_at": "2026-03-19T12:41:29.232499Z",
-  "sessions": [
-    {
-      "sessionId": "agent:architect-agent:subagent:576757a3-3e3f-40b8-95e3-b91cfeecc038",
-      "agentId": "architect-agent",
-      "phase": "analysis",
-      "label": "SA-legacy-anatomy-retry",
-      "taskSummary": "存量代码摸排 openclaw-agent-dashboard",
-      "spawnedAt": "2026-03-19T12:31:15.960549Z",
-      "lastActivityAt": "2026-03-19T12:31:15.960549Z",
-      "completedAt": "2026-03-19T12:33:40.413958Z",
-      "status": "COMPLETED",
-      "failureType": null,
-      "retryCount": 0,
-      "retryHistory": [],
-      "expectedArtifacts": [
-        "/home/ubuntu/vrt-projects/projects/openclaw-agent-dashboard/.staging/legacy_code_anatomy.md"
-      ],
-      "actualArtifacts": [
-        "/home/ubuntu/vrt-projects/projects/openclaw-agent-dashboard/.staging/legacy_code_anatomy.md"
-      ],
-      "parentId": null,
-      "lastError": null
-    },
-    {
-      "sessionId": "agent:analyst-agent:subagent:a0a08020-d544-4695-9efe-24d5948cd928",
-      "agentId": "analyst-agent",
-      "phase": "requirements",
-      "label": "BA-PRD-version-display",
-      "taskSummary": "编写界面显示版本号PRD",
-      "spawnedAt": "2026-03-19T12:34:29.817387Z",
-      "lastActivityAt": "2026-03-19T12:34:29.817387Z",
-      "completedAt": "2026-03-19T12:41:29.232476Z",
-      "status": "COMPLETED",
-      "failureType": null,
-      "retryCount": 0,
-      "retryHistory": [],
-      "expectedArtifacts": [
-        "/home/ubuntu/vrt-projects/projects/openclaw-agent-dashboard/.staging/specs/VERSION_DISPLAY_spec.md"
-      ],
-      "actualArtifacts": [
-        "/home/ubuntu/vrt-projects/projects/openclaw-agent-dashboard/.staging/specs/VERSION_DISPLAY_spec.md"
-      ],
-      "parentId": null,
-      "lastError": null
-    }
-  ],
-  "statistics": {
-    "total": 2,
-    "running": 0,
-    "completed": 2,
-    "stuck": 0,
-    "failed": 0
-  }
-}