Fix frontend build dependencies

- Migrate OpenClaw from HTTP (port 8004) to WebSocket (port 18789)
- Add workspace file list and content preview handlers
- Add OpenClawStatus component with agent/skills view
- Add OpenClawView panel in trader interface
- Add Zustand store for OpenClaw state management
- Fix gateway logging noise (yfinance, websockets)
- Fix RunWorkspaceManager.get_agent_asset_dir attribute error
- Handle missing workspace files gracefully in preview

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.env.example

@@ -0,0 +1,41 @@
+# Copy this file to `.env` for local development.
+# Keep `.env` untracked and never paste real secrets into tracked files.
+
+# ================== General Configuration | 通用配置 ==================
+TICKERS=AAPL,MSFT,GOOGL,AMZN,NVDA,META,TSLA,AMD,NFLX,AVGO,PLTR,COIN
+
+# Financial Data API
+# At least `FINANCIAL_DATASETS_API_KEY` is required when using `FIN_DATA_SOURCE=financial_datasets`.
+# `FINNHUB_API_KEY` is recommended for `FIN_DATA_SOURCE=finnhub` and required for live mode.
+FIN_DATA_SOURCE=finnhub
+ENABLED_DATA_SOURCES=financial_datasets,finnhub,yfinance,local_csv
+FINANCIAL_DATASETS_API_KEY=
+FINNHUB_API_KEY=
+POLYGON_API_KEY=
+MARKET_DB_PATH=
+
+# Model API
+OPENAI_API_KEY=
+OPENAI_BASE_URL=
+MODEL_NAME=qwen3-max-preview
+EXPLAIN_ENRICH_USE_LLM=false
+EXPLAIN_ENRICH_MODEL_PROVIDER=
+EXPLAIN_ENRICH_MODEL_NAME=
+EXPLAIN_RANGE_USE_LLM=
+
+# Memory module
+MEMORY_API_KEY=
+
+# ================== Agent-Specific Model Configuration | Agent特定模型配置 ==================
+AGENT_SENTIMENT_ANALYST_MODEL_NAME=deepseek-v3.2-exp
+AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4.6
+AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=qwen3-max-preview
+AGENT_VALUATION_ANALYST_MODEL_NAME=Moonshot-Kimi-K2-Instruct
+AGENT_RISK_MANAGER_MODEL_NAME=qwen3-max-preview
+AGENT_PORTFOLIO_MANAGER_MODEL_NAME=qwen3-max-preview
+
+# ================== Advanced Configuration | 高阶配置 ==================
+MAX_COMM_CYCLES=2
+MARGIN_REQUIREMENT=0.5
+DATA_START_DATE=2022-01-01
+AUTO_UPDATE_DATA=true

.gitignore

@@ -51,13 +51,15 @@ node_modules
 outputs/
 /production/
 /smoke_test/
-/smoke_live_mock/
 
 # Local tooling state
-/.omc/
+.omc/
 /.pydeps/
 /referance/
 
+# Run outputs
+/runs/
+
 # Data files
 backend/data/ret_data/
 

.omc/project-memory.json

@@ -1,6 +1,6 @@
 {
   "version": "1.0.0",
-  "lastScanned": 1773938154948,
+  "lastScanned": 1774515151036,
   "projectRoot": "/Users/cillin/workspeace/evotraders",
   "techStack": {
     "languages": [
@@ -11,14 +11,6 @@
         "markers": [
           "pyproject.toml"
         ]
-      },
-      {
-        "name": "C/C++",
-        "version": null,
-        "confidence": "high",
-        "markers": [
-          "Makefile"
-        ]
       }
     ],
     "frameworks": [
@@ -32,8 +24,8 @@
     "runtime": null
   },
   "build": {
-    "buildCommand": "make build",
+    "buildCommand": null,
-    "testCommand": "make test",
+    "testCommand": "pytest",
     "lintCommand": "ruff check",
     "devCommand": null,
     "scripts": {}
@@ -58,24 +50,13 @@
   },
   "customNotes": [],
   "directoryMap": {
-    "agent-service": {
-      "path": "agent-service",
-      "purpose": null,
-      "fileCount": 2,
-      "lastAccessed": 1773938154941,
-      "keyFiles": [
-        "Dockerfile",
-        "requirements.txt"
-      ]
-    },
     "backend": {
       "path": "backend",
       "purpose": null,
-      "fileCount": 5,
+      "fileCount": 4,
-      "lastAccessed": 1773938154941,
+      "lastAccessed": 1774515151025,
       "keyFiles": [
         "__init__.py",
-        "app.py",
         "cli.py",
         "gateway_server.py",
         "main.py"
@@ -85,37 +66,41 @@
       "path": "backtest",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154941,
+      "lastAccessed": 1774515151026,
       "keyFiles": []
     },
     "data": {
       "path": "data",
       "purpose": "Data files",
-      "fileCount": 1,
+      "fileCount": 3,
-      "lastAccessed": 1773938154941,
+      "lastAccessed": 1774515151027,
       "keyFiles": [
-        "market_research.db"
+        "market_research.db",
+        "market_research.db-shm",
+        "market_research.db-wal"
       ]
     },
     "deploy": {
       "path": "deploy",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154942,
+      "lastAccessed": 1774515151027,
       "keyFiles": []
     },
     "docs": {
       "path": "docs",
       "purpose": "Documentation",
-      "fileCount": 0,
+      "fileCount": 1,
-      "lastAccessed": 1773938154942,
+      "lastAccessed": 1774515151027,
-      "keyFiles": []
+      "keyFiles": [
+        "compat-removal-plan.md"
+      ]
     },
     "evotraders.egg-info": {
       "path": "evotraders.egg-info",
       "purpose": null,
       "fileCount": 6,
-      "lastAccessed": 1773938154942,
+      "lastAccessed": 1774515151028,
       "keyFiles": [
         "PKG-INFO",
         "SOURCES.txt",
@@ -128,7 +113,7 @@
       "path": "frontend",
       "purpose": null,
       "fileCount": 13,
-      "lastAccessed": 1773938154942,
+      "lastAccessed": 1774515151028,
       "keyFiles": [
         "README.md",
         "components.json",
@@ -141,51 +126,28 @@
       "path": "live",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154943,
+      "lastAccessed": 1774515151028,
       "keyFiles": []
     },
-    "logs": {
-      "path": "logs",
-      "purpose": null,
-      "fileCount": 7,
-      "lastAccessed": 1773938154943,
-      "keyFiles": [
-        "2026-03-16_00-48-03.log",
-        "2026-03-18_23-17-29.log",
-        "2026-03-18_23-17-30.2026-03-18_23-17-30_000801.log.zip",
-        "2026-03-18_23-17-30.log",
-        "2026-03-19_00-18-04.log"
-      ]
-    },
-    "news-service": {
-      "path": "news-service",
-      "purpose": null,
-      "fileCount": 3,
-      "lastAccessed": 1773938154943,
-      "keyFiles": [
-        "Dockerfile",
-        "requirements.txt"
-      ]
-    },
     "reference": {
       "path": "reference",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154943,
+      "lastAccessed": 1774515151028,
       "keyFiles": []
     },
     "runs": {
       "path": "runs",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151029,
       "keyFiles": []
     },
     "scripts": {
       "path": "scripts",
       "purpose": "Build/utility scripts",
       "fileCount": 1,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151030,
       "keyFiles": [
         "run_prod.sh"
       ]
@@ -194,7 +156,7 @@
       "path": "services",
       "purpose": "Business logic services",
       "fileCount": 1,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151030,
       "keyFiles": [
         "README.md"
       ]
@@ -203,43 +165,14 @@
       "path": "shared",
       "purpose": null,
       "fileCount": 0,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151030,
       "keyFiles": []
     },
-    "trading-service": {
-      "path": "trading-service",
-      "purpose": null,
-      "fileCount": 4,
-      "lastAccessed": 1773938154944,
-      "keyFiles": [
-        "Dockerfile",
-        "README.md",
-        "requirements.txt"
-      ]
-    },
-    "workspaces": {
-      "path": "workspaces",
-      "purpose": null,
-      "fileCount": 0,
-      "lastAccessed": 1773938154944,
-      "keyFiles": []
-    },
-    "agent-service/src": {
-      "path": "agent-service/src",
-      "purpose": "Source code",
-      "fileCount": 5,
-      "lastAccessed": 1773938154944,
-      "keyFiles": [
-        "__init__.py",
-        "config.py",
-        "main.py"
-      ]
-    },
     "backend/api": {
       "path": "backend/api",
       "purpose": "API routes",
       "fileCount": 5,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151030,
       "keyFiles": [
         "__init__.py",
         "agents.py",
@@ -250,7 +183,7 @@
       "path": "backend/config",
       "purpose": "Configuration files",
       "fileCount": 6,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151030,
       "keyFiles": [
         "__init__.py",
         "agent_profiles.yaml",
@@ -260,8 +193,8 @@
     "backend/data": {
       "path": "backend/data",
       "purpose": "Data files",
-      "fileCount": 13,
+      "fileCount": 12,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151031,
       "keyFiles": [
         "__init__.py",
         "cache.py",
@@ -272,7 +205,7 @@
       "path": "docs/assets",
       "purpose": "Static assets",
       "fileCount": 5,
-      "lastAccessed": 1773938154944,
+      "lastAccessed": 1774515151031,
       "keyFiles": [
         "dashboard.jpg",
         "evotraders_demo.gif",
@@ -283,7 +216,7 @@
       "path": "frontend/dist",
       "purpose": "Distribution/build output",
       "fileCount": 2,
-      "lastAccessed": 1773938154945,
+      "lastAccessed": 1774515151031,
       "keyFiles": [
         "index.html",
         "trading_logo.png"
@@ -293,331 +226,309 @@
       "path": "frontend/node_modules",
       "purpose": "Dependencies",
       "fileCount": 1,
-      "lastAccessed": 1773938154947,
+      "lastAccessed": 1774515151036,
       "keyFiles": []
-    },
-    "news-service/src": {
-      "path": "news-service/src",
-      "purpose": "Source code",
-      "fileCount": 3,
-      "lastAccessed": 1773938154948,
-      "keyFiles": [
-        "__init__.py",
-        "config.py",
-        "main.py"
-      ]
-    },
-    "trading-service/src": {
-      "path": "trading-service/src",
-      "purpose": "Source code",
-      "fileCount": 8,
-      "lastAccessed": 1773938154948,
-      "keyFiles": [
-        "__init__.py",
-        "config.py",
-        "main.py"
-      ]
     }
   },
   "hotPaths": [
     {
-      "path": "backend/agents/factory.py",
+      "path": "frontend/src/hooks/useWebSocketConnection.js",
-      "accessCount": 17,
+      "accessCount": 100,
-      "lastAccessed": 1773939950376,
+      "lastAccessed": 1774550862686,
-      "type": "file"
-    },
-    {
-      "path": "backend",
-      "accessCount": 16,
-      "lastAccessed": 1773940042371,
-      "type": "directory"
-    },
-    {
-      "path": "",
-      "accessCount": 13,
-      "lastAccessed": 1773939899611,
-      "type": "directory"
-    },
-    {
-      "path": "backend/main.py",
-      "accessCount": 7,
-      "lastAccessed": 1773939993951,
-      "type": "file"
-    },
-    {
-      "path": "backend/gateway_server.py",
-      "accessCount": 7,
-      "lastAccessed": 1773940004402,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/main.py",
-      "accessCount": 5,
-      "lastAccessed": 1773938385662,
-      "type": "file"
-    },
-    {
-      "path": "backend/core/pipeline.py",
-      "accessCount": 5,
-      "lastAccessed": 1773940024933,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/enrich/news_enricher.py",
-      "accessCount": 4,
-      "lastAccessed": 1773938508417,
-      "type": "file"
-    },
-    {
-      "path": "start-dev.sh",
-      "accessCount": 4,
-      "lastAccessed": 1773939259381,
-      "type": "file"
-    },
-    {
-      "path": "services/README.md",
-      "accessCount": 4,
-      "lastAccessed": 1773939281935,
-      "type": "file"
-    },
-    {
-      "path": "backend/app.py",
-      "accessCount": 4,
-      "lastAccessed": 1773939648215,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/routes/news.py",
-      "accessCount": 3,
-      "lastAccessed": 1773938438928,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news",
-      "accessCount": 3,
-      "lastAccessed": 1773938468730,
-      "type": "directory"
-    },
-    {
-      "path": "frontend/src/config/constants.js",
-      "accessCount": 3,
-      "lastAccessed": 1773939204395,
       "type": "file"
     },
     {
       "path": "backend/services/gateway.py",
-      "accessCount": 3,
+      "accessCount": 98,
-      "lastAccessed": 1773939672930,
+      "lastAccessed": 1774550272354,
       "type": "file"
     },
     {
-      "path": "backend/core/__init__.py",
+      "path": "backend/services/gateway_openclaw_handlers.py",
-      "accessCount": 3,
+      "accessCount": 91,
-      "lastAccessed": 1773939963627,
+      "lastAccessed": 1774550256325,
       "type": "file"
     },
     {
-      "path": "backend/services/trading/main.py",
+      "path": "backend/api/openclaw.py",
-      "accessCount": 2,
+      "accessCount": 48,
-      "lastAccessed": 1773938360736,
+      "lastAccessed": 1774545375555,
       "type": "file"
     },
     {
-      "path": "backend/services/agents/main.py",
+      "path": "frontend/src/hooks/useOpenClawPanel.js",
-      "accessCount": 2,
+      "accessCount": 42,
-      "lastAccessed": 1773938361040,
+      "lastAccessed": 1774550688926,
       "type": "file"
     },
     {
-      "path": "backend/services/trading/data/__init__.py",
+      "path": "shared/client/openclaw_client.py",
-      "accessCount": 2,
+      "accessCount": 39,
-      "lastAccessed": 1773938402496,
+      "lastAccessed": 1774545484770,
       "type": "file"
     },
     {
-      "path": "backend/services/news/explain/__init__.py",
+      "path": "frontend/src",
-      "accessCount": 2,
+      "accessCount": 35,
-      "lastAccessed": 1773938460019,
+      "lastAccessed": 1774550715529,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/enrich/__init__.py",
-      "accessCount": 2,
-      "lastAccessed": 1773938465216,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/explain/range_explainer.py",
-      "accessCount": 2,
-      "lastAccessed": 1773938481152,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/enrich/llm_enricher.py",
-      "accessCount": 2,
-      "lastAccessed": 1773938499885,
-      "type": "file"
-    },
-    {
-      "path": "CLAUDE.md",
-      "accessCount": 2,
-      "lastAccessed": 1773939273598,
-      "type": "file"
-    },
-    {
-      "path": "backend/agents/__init__.py",
-      "accessCount": 2,
-      "lastAccessed": 1773939883015,
-      "type": "file"
-    },
-    {
-      "path": "backend/agents/agent_core.py",
-      "accessCount": 2,
-      "lastAccessed": 1773939886997,
-      "type": "file"
-    },
-    {
-      "path": "Makefile",
-      "accessCount": 1,
-      "lastAccessed": 1773938226307,
-      "type": "file"
-    },
-    {
-      "path": "docker-compose.yml",
-      "accessCount": 1,
-      "lastAccessed": 1773938226360,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/news/shared/trading_client.py",
-      "accessCount": 1,
-      "lastAccessed": 1773938370618,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/agents",
-      "accessCount": 1,
-      "lastAccessed": 1773938397772,
       "type": "directory"
     },
     {
-      "path": "backend/services/trading",
+      "path": "reference/openclaw/src",
-      "accessCount": 1,
+      "accessCount": 33,
-      "lastAccessed": 1773938397823,
+      "lastAccessed": 1774550840611,
       "type": "directory"
     },
     {
-      "path": "backend/services",
+      "path": "backend/services/openclaw_cli.py",
-      "accessCount": 1,
+      "accessCount": 31,
-      "lastAccessed": 1773938405541,
+      "lastAccessed": 1774545484887,
-      "type": "directory"
-    },
-    {
-      "path": "backend/services/news/config.py",
-      "accessCount": 1,
-      "lastAccessed": 1773938638664,
       "type": "file"
     },
     {
-      "path": "shared/client/news_client.py",
+      "path": "frontend/src/components/TraderView.jsx",
-      "accessCount": 1,
+      "accessCount": 23,
-      "lastAccessed": 1773938638715,
+      "lastAccessed": 1774543366574,
       "type": "file"
     },
     {
-      "path": "shared/client/trading_client.py",
+      "path": "shared/models/openclaw.py",
-      "accessCount": 1,
+      "accessCount": 22,
-      "lastAccessed": 1773938638770,
+      "lastAccessed": 1774545419541,
       "type": "file"
     },
     {
-      "path": "backend/api",
+      "path": "frontend/src/store/openclawStore.js",
-      "accessCount": 1,
+      "accessCount": 20,
-      "lastAccessed": 1773938669143,
+      "lastAccessed": 1774550319533,
-      "type": "directory"
+      "type": "file"
     },
     {
-      "path": "frontend",
+      "path": "frontend/src/App.jsx",
-      "accessCount": 1,
+      "accessCount": 18,
-      "lastAccessed": 1773938669195,
+      "lastAccessed": 1774544542524,
-      "type": "directory"
-    },
-    {
-      "path": ".env.example",
-      "accessCount": 1,
-      "lastAccessed": 1773938849397,
       "type": "file"
     },
     {
       "path": "frontend/src/services/websocket.js",
-      "accessCount": 1,
+      "accessCount": 18,
-      "lastAccessed": 1773938849448,
+      "lastAccessed": 1774549669596,
+      "type": "file"
+    },
+    {
+      "path": "start-dev.sh",
+      "accessCount": 15,
+      "lastAccessed": 1774548224246,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/components/RuntimeView.jsx",
+      "accessCount": 14,
+      "lastAccessed": 1774518525793,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/components/AppShell.jsx",
+      "accessCount": 13,
+      "lastAccessed": 1774533781725,
+      "type": "file"
+    },
+    {
+      "path": "backend/main.py",
+      "accessCount": 13,
+      "lastAccessed": 1774548236340,
+      "type": "directory"
+    },
+    {
+      "path": "backend/apps/openclaw_service.py",
+      "accessCount": 10,
+      "lastAccessed": 1774547900186,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/components/OpenClawStatusPanel.jsx",
+      "accessCount": 8,
+      "lastAccessed": 1774533622019,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/commands",
+      "accessCount": 7,
+      "lastAccessed": 1774530402019,
+      "type": "directory"
+    },
+    {
+      "path": "frontend/src/config/constants.js",
+      "accessCount": 7,
+      "lastAccessed": 1774544689658,
+      "type": "file"
+    },
+    {
+      "path": "",
+      "accessCount": 6,
+      "lastAccessed": 1774550700047,
+      "type": "directory"
+    },
+    {
+      "path": "backend/services",
+      "accessCount": 5,
+      "lastAccessed": 1774550692490,
+      "type": "directory"
+    },
+    {
+      "path": "frontend/src/store/uiStore.js",
+      "accessCount": 4,
+      "lastAccessed": 1774533747700,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/styles/GlobalStyles.jsx",
+      "accessCount": 4,
+      "lastAccessed": 1774533753657,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/store/agentStore.js",
+      "accessCount": 3,
+      "lastAccessed": 1774517930592,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/cli/skills-cli.ts",
+      "accessCount": 3,
+      "lastAccessed": 1774527140107,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/commands/agents.commands.list.ts",
+      "accessCount": 3,
+      "lastAccessed": 1774533427441,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/store/runtimeStore.js",
+      "accessCount": 2,
+      "lastAccessed": 1774517930660,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/hooks/useAgentWorkspacePanel.js",
+      "accessCount": 2,
+      "lastAccessed": 1774518021290,
       "type": "file"
     },
     {
       "path": "frontend/src/services/runtimeApi.js",
-      "accessCount": 1,
+      "accessCount": 2,
-      "lastAccessed": 1773938849500,
+      "lastAccessed": 1774518025465,
       "type": "file"
     },
     {
-      "path": "backend/services/agents/routes/websocket.py",
+      "path": "reference/openclaw/src/commands/agents.commands.delete.ts",
-      "accessCount": 1,
+      "accessCount": 2,
-      "lastAccessed": 1773939001692,
+      "lastAccessed": 1774530389553,
       "type": "file"
     },
     {
-      "path": "backend/services/agents/routes/agents.py",
+      "path": "reference/openclaw/src/commands/agents.commands.add.ts",
-      "accessCount": 1,
+      "accessCount": 2,
-      "lastAccessed": 1773939016291,
+      "lastAccessed": 1774530389605,
-      "type": "file"
-    },
-    {
-      "path": "backend/services/agents/routes/run.py",
-      "accessCount": 1,
-      "lastAccessed": 1773939016343,
-      "type": "file"
-    },
-    {
-      "path": "backend/__init__.py",
-      "accessCount": 1,
-      "lastAccessed": 1773939648323,
       "type": "file"
     },
     {
       "path": "backend/api/__init__.py",
-      "accessCount": 1,
+      "accessCount": 2,
-      "lastAccessed": 1773939658650,
+      "lastAccessed": 1774542416191,
       "type": "file"
     },
     {
-      "path": "backend/runtime/__init__.py",
+      "path": "frontend/vite.config.js",
-      "accessCount": 1,
+      "accessCount": 2,
-      "lastAccessed": 1773939658687,
+      "lastAccessed": 1774544772960,
       "type": "file"
     },
     {
-      "path": "backend/agents/base/evo_agent.py",
+      "path": "frontend/src/store/index.js",
       "accessCount": 1,
-      "lastAccessed": 1773939664916,
+      "lastAccessed": 1774515811752,
       "type": "file"
     },
     {
-      "path": "backend/agents/analyst.py",
+      "path": "frontend/src/store/marketStore.js",
       "accessCount": 1,
-      "lastAccessed": 1773939664967,
+      "lastAccessed": 1774515838923,
       "type": "file"
     },
     {
-      "path": "backend/agents/base/hooks.py",
+      "path": "frontend/src/store/portfolioStore.js",
       "accessCount": 1,
-      "lastAccessed": 1773939672727,
+      "lastAccessed": 1774515839687,
       "type": "file"
     },
     {
-      "path": "pyproject.toml",
+      "path": "frontend/src/index.css",
       "accessCount": 1,
-      "lastAccessed": 1773939672778,
+      "lastAccessed": 1774515988837,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/App.css",
+      "accessCount": 1,
+      "lastAccessed": 1774515998423,
+      "type": "file"
+    },
+    {
+      "path": "frontend/package.json",
+      "accessCount": 1,
+      "lastAccessed": 1774516005569,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/hooks/useAgentDataRequests.js",
+      "accessCount": 1,
+      "lastAccessed": 1774517930219,
+      "type": "file"
+    },
+    {
+      "path": "backend/services/gateway_admin_handlers.py",
+      "accessCount": 1,
+      "lastAccessed": 1774517937966,
+      "type": "file"
+    },
+    {
+      "path": "backend/apps/agent_service.py",
+      "accessCount": 1,
+      "lastAccessed": 1774517946208,
+      "type": "file"
+    },
+    {
+      "path": "frontend/src/hooks",
+      "accessCount": 1,
+      "lastAccessed": 1774517946260,
+      "type": "directory"
+    },
+    {
+      "path": "frontend/src/hooks/useFeedProcessor.js",
+      "accessCount": 1,
+      "lastAccessed": 1774517952115,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/commands/models/set.ts",
+      "accessCount": 1,
+      "lastAccessed": 1774526963526,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/commands/models/list.ts",
+      "accessCount": 1,
+      "lastAccessed": 1774526963632,
+      "type": "file"
+    },
+    {
+      "path": "reference/openclaw/src/cli/skills-cli.format.ts",
+      "accessCount": 1,
+      "lastAccessed": 1774526963684,
       "type": "file"
     }
   ],

.omc/state/hud-state.json

@@ -1,6 +1,6 @@
 {
-  "timestamp": "2026-03-19T16:36:52.471Z",
+  "timestamp": "2026-03-27T04:53:52.906Z",
   "backgroundTasks": [],
-  "sessionStartTimestamp": "2026-03-19T16:36:42.224Z",
+  "sessionStartTimestamp": "2026-03-27T04:53:21.944Z",
-  "sessionId": "ef02339a-1eec-4c7a-95ac-c8cfa0b5067d"
+  "sessionId": "cbb9004e-771b-4e82-95d4-cea6d9753642"
 }

.omc/state/hud-stdin-cache.json

@@ -1 +1 @@
-{"session_id":"ef02339a-1eec-4c7a-95ac-c8cfa0b5067d","transcript_path":"/Users/cillin/.claude/projects/-Users-cillin-workspeace-evotraders/ef02339a-1eec-4c7a-95ac-c8cfa0b5067d.jsonl","cwd":"/Users/cillin/workspeace/evotraders","model":{"id":"MiniMax-M2.7-highspeed","display_name":"MiniMax-M2.7-highspeed"},"workspace":{"current_dir":"/Users/cillin/workspeace/evotraders","project_dir":"/Users/cillin/workspeace/evotraders","added_dirs":[]},"version":"2.1.78","output_style":{"name":"default"},"cost":{"total_cost_usd":17.458779250000003,"total_duration_ms":1866224,"total_api_duration_ms":1188013,"total_lines_added":257,"total_lines_removed":290},"context_window":{"total_input_tokens":195204,"total_output_tokens":48917,"context_window_size":200000,"current_usage":{"input_tokens":481,"output_tokens":0,"cache_creation_input_tokens":149,"cache_read_input_tokens":163286},"used_percentage":82,"remaining_percentage":18},"exceeds_200k_tokens":false}
+{"session_id":"cbb9004e-771b-4e82-95d4-cea6d9753642","transcript_path":"/Users/cillin/.claude/projects/-Users-cillin-workspeace-evotraders/cbb9004e-771b-4e82-95d4-cea6d9753642.jsonl","cwd":"/Users/cillin/workspeace/evotraders","model":{"id":"MiniMax-M2.7-highspeed","display_name":"MiniMax-M2.7-highspeed"},"workspace":{"current_dir":"/Users/cillin/workspeace/evotraders","project_dir":"/Users/cillin/workspeace/evotraders","added_dirs":[]},"version":"2.1.78","output_style":{"name":"default"},"cost":{"total_cost_usd":0.660433,"total_duration_ms":168502,"total_api_duration_ms":37670,"total_lines_added":0,"total_lines_removed":0},"context_window":{"total_input_tokens":14416,"total_output_tokens":1705,"context_window_size":200000,"current_usage":{"input_tokens":461,"output_tokens":214,"cache_creation_input_tokens":0,"cache_read_input_tokens":53991},"used_percentage":27,"remaining_percentage":73},"exceeds_200k_tokens":false}

.omc/state/idle-notif-cooldown.json

@@ -1,3 +1,3 @@
 {
-  "lastSentAt": "2026-03-19T17:02:32.170Z"
+  "lastSentAt": "2026-03-27T04:55:49.635Z"
 }

.omc/state/subagent-tracking.json

@@ -1,17 +0,0 @@
-{
-  "agents": [
-    {
-      "agent_id": "a8305a91e192b2196",
-      "agent_type": "Explore",
-      "started_at": "2026-03-19T17:00:33.284Z",
-      "parent_mode": "none",
-      "status": "completed",
-      "completed_at": "2026-03-19T17:02:19.439Z",
-      "duration_ms": 106155
-    }
-  ],
-  "total_spawned": 1,
-  "total_completed": 1,
-  "total_failed": 0,
-  "last_updated": "2026-03-19T17:02:39.175Z"
-}

CLAUDE.md

@@ -1,10 +1,12 @@
 # CLAUDE.md
 
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
 本文件为 Claude Code (claude.ai/code) 在此代码库中工作时提供指导。
 
 ## 项目概述
 
-EvoTraders 是一个自进化多智能体交易系统，由 6 个 AI Agent（4 名分析师 + 投资经理 + 风控经理）协作完成交易决策。Agent 基于 AgentScope 框架构建，配合 ReMe 记忆系统实现持续学习。
+大时代 是一个自进化多智能体交易系统，由 6 个 AI Agent（4 名分析师 + 投资经理 + 风控经理）协作完成交易决策。Agent 基于 AgentScope 框架构建，配合 ReMe 记忆系统实现持续学习。
 
 ## 常用命令
 
@@ -18,23 +20,24 @@ uv pip install -e .
 evotraders backtest --start 2025-11-01 --end 2025-12-01  # 回测模式
 evotraders backtest --start 2025-11-01 --end 2025-12-01 --enable-memory  # 带记忆回测
 evotraders live                              # 实盘交易
-evotraders live --mock                       # 模拟/测试模式
 evotraders live -t 22:30                     # 定时每日交易
 evotraders frontend                          # 启动可视化界面
 
 # 开发服务器
-./start-dev.sh  # 启动全部 4 个微服务
+./start-dev.sh  # 启动全部 4 个微服务 (agent, runtime, trading, news)
 
-# 单独启动某个服务
+# Gateway WebSocket 服务器
-python -m uvicorn backend.apps.agent_service:app --host 0.0.0.0 --port 8000 --reload
+python backend/main.py --mode live --config-name live
+
+# 单独启动微服务
 python -m uvicorn backend.apps.runtime_service:app --host 0.0.0.0 --port 8003 --reload
+python -m uvicorn backend.apps.agent_service:app --host 0.0.0.0 --port 8000 --reload
 python -m uvicorn backend.apps.trading_service:app --host 0.0.0.0 --port 8001 --reload
 python -m uvicorn backend.apps.news_service:app --host 0.0.0.0 --port 8002 --reload
 
 # 测试
 pytest backend/tests                          # 运行全部测试
-pytest backend/tests/test_news_service_app.py -v  # 运行单个测试文件
+pytest backend/tests/test_news_service_app.py -v  # 运行单个测试
-pytest backend/tests/test_news_service_app.py::test_news_service_routes_are_exposed -v  # 运行单个测试
 ```
 
 ### Frontend (React)
@@ -46,142 +49,236 @@ npm run build        # 生产构建
 npm run lint         # ESLint 检查
 npm run lint:fix     # ESLint 自动修复
 npm run test         # Vitest 单元测试
-npm run test:watch   # 监听模式
 ```
 
 ## 架构概览
 
-### 微服务架构 (`backend/apps/`)
+### 系统分层
 
-项目采用 split-first 微服务架构，4 个独立的 FastAPI 服务：
+```
-
+┌─────────────────────────────────────────────────────────────┐
-| 服务 | 入口 | 端口 | 职责 |
+│                     Frontend (React)                        │
-|------|------|------|------|
+│         WebSocket ws://localhost:8765 连接 Gateway           │
-| agent_service | `backend.apps.agent_service:app` | 8000 | Agent 生命周期、工作区管理 |
+└─────────────────────────────────────────────────────────────┘
-| runtime_service | `backend.apps.runtime_service:app` | 8003 | 运行时配置、任务启动 |
+                              │
-| trading_service | `backend.apps.trading_service:app` | 8001 | 市场数据、交易操作 |
+                              ▼
-| news_service | `backend.apps.news_service:app` | 8002 | 新闻、新闻富化、解释功能 |
+┌─────────────────────────────────────────────────────────────┐
-
+│              Gateway (backend/services/gateway.py)           │
-服务间通过环境变量通信（详见 `start-dev.sh`）：
+│         WebSocket 服务器，编排 Pipeline，4 阶段启动            │
-```bash
+└─────────────────────────────────────────────────────────────┘
-export TRADING_SERVICE_URL=http://localhost:8001
+          │               │               │               │
-export NEWS_SERVICE_URL=http://localhost:8002
+          ▼               ▼               ▼               ▼
-export RUNTIME_SERVICE_URL=http://localhost:8003
+┌────────────┐  ┌────────────┐  ┌────────────┐  ┌────────────┐
+│   Market   │  │  Storage   │  │  Pipeline  │  │ Scheduler  │
+│  Service   │  │  Service   │  │            │  │            │
+└────────────┘  └────────────┘  └────────────┘  └────────────┘
+                                           │
+                    ┌──────────────────────┼──────────────────────┐
+                    ▼                      ▼                      ▼
+              ┌──────────┐         ┌──────────┐         ┌──────────┐
+              │ Analysts │         │    PM    │         │   Risk   │
+              │  (4 个)  │         │          │         │ Manager  │
+              └──────────┘         └──────────┘         └──────────┘
 ```
 
-### Gateway 网关 (`backend/services/gateway.py`)
+### 微服务架构 (`backend/apps/`)
 
-Gateway 是统一的请求路由器，根据路径前缀将请求转发到对应的微服务：
+| 服务 | 端口 | 职责 |
-- `/control/*` → agent_service
+|------|------|------|
-- `/runtime/*` → runtime_service
+| runtime_service | 8003 | 运行时配置、任务启动、Pipeline Runner |
-- `/trading/*` → trading_service
+| agent_service | 8000 | Agent 生命周期、工作区管理 |
-- `/news/*` → news_service
+| trading_service | 8001 | 市场数据、交易操作 |
+| news_service | 8002 | 新闻、新闻富化、解释功能 |
 
-新增接口时应注册到对应的 service app，而非直接添加到 gateway。
+### Gateway 4 阶段启动 (`backend/services/gateway.py`)
 
-### 共享客户端 (`shared/client/`)
+1. **WebSocket Server** - 前端立即可连接
+2. **Market Service** - 价格数据开始推送
+3. **Market Status Monitor** - 市场状态监控
+4. **Scheduler** - 交易周期开始
 
-统一的服务客户端库，所有前端和后端服务间通信都使用此处定义的客户端：
+### 运行时管理层 (`backend/runtime/`)
 
-| 客户端 | 用途 |
+| 文件 | 职责 |
-|--------|------|
+|------|------|
-| `ControlPlaneClient` | Agent 服务通信 |
+| `manager.py` | TradingRuntimeManager - 全局运行时管理器，agent 注册、会话、事件快照 |
-| `RuntimeServiceClient` | 运行时服务通信 |
+| `agent_runtime.py` | AgentRuntimeState - 单 agent 状态（status、last_session） |
-| `TradingServiceClient` | 交易服务通信 |
+| `context.py` | TradingRunContext - 运行上下文 |
-| `NewsServiceClient` | 新闻服务通信 |
+| `session.py` | TradingSessionKey - 交易日会话键 |
+| `registry.py` | RuntimeRegistry - agent 状态注册表 |
 
-### 领域层 (`backend/domains/`)
+快照持久化到 `runs/<run_id>/state/runtime_state.json`。
 
-业务逻辑按领域分离：
+### Pipeline 执行 (`backend/core/`)
 
-- `news.py` - 新闻领域操作
+| 文件 | 职责 |
-- `trading.py` - 交易领域操作
+|------|------|
+| `pipeline.py` | TradingPipeline - 核心编排器（分析→沟通→决策→执行→评估） |
+| `pipeline_runner.py` | REST API 触发的独立执行，5 阶段启动 |
+| `scheduler.py` | BacktestScheduler、Scheduler - 回测/实盘调度 |
+| `state_sync.py` | StateSync - 状态同步和广播 |
 
 ## 后端结构
 
 ```
 backend/
 ├── agents/              # 多智能体实现
-│   ├── base/         # 核心类、Hooks、评估
+│   ├── analyst.py           # AnalystAgent 基类
-│   │   ├── evo_agent.py        # 基于 AgentScope 的核心实现
+│   ├── portfolio_manager.py # PMAgent 投资经理
-│   │   ├── hooks.py             # 生命周期 Hooks
+│   ├── risk_manager.py      # RiskAgent 风控经理
-│   │   │   ├── BootstrapHook           # 启动初始化
-│   │   │   ├── MemoryCompactionHook    # 内存压缩（基于 CoPaw）
-│   │   │   ├── HeartbeatHook           # 心跳检测
-│   │   │   └── WorkspaceWatchHook      # 工作区监控
-│   │   ├── evaluation_hook.py   # 执行后评估
-│   │   ├── skill_adaptation_hook.py  # 动态技能适配
-│   │   └── tool_guard.py        # 工具调用守卫
-│   ├── prompts/      # Agent 提示词和角色定义
-│   │   ├── analyst/personas.yaml  # 分析师角色配置
-│   │   └── portfolio_manager/
-│   ├── team/         # 团队协作逻辑
-│   │   ├── registry.py        # Agent 注册表
-│   │   ├── coordinator.py     # 协作协调器
-│   │   ├── messenger.py       # 消息传递
-│   │   └── task_delegator.py  # 任务分发
 │   ├── factory.py           # Agent 实例工厂
-│   ├── skills_manager.py  # 技能加载管理（6 种作用域）
+│   ├── toolkit_factory.py   # 工具集工厂
-│   └── toolkit_factory.py  # 工具集工厂
+│   ├── skills_manager.py    # 技能加载管理
-├── apps/             # 微服务入口（split-first）
+│   ├── workspace_manager.py # 工作区管理
-│   ├── agent_service.py
+│   ├── skill_loader.py      # 技能加载器
-│   ├── runtime_service.py
+│   ├── agent_workspace.py   # Agent 工作区
-│   ├── trading_service.py
+│   ├── prompt_loader.py     # Prompt 加载器
-│   └── news_service.py
+│   ├── prompt_factory.py    # Prompt 工厂
+│   ├── skill_metadata.py    # 技能元数据
+│   ├── registry.py          # Agent 注册表
+│   ├── team_pipeline_config.py  # 团队 Pipeline 配置
+│   ├── compat.py            # 兼容性层
+│   ├── templates.py         # 模板
+│   ├── workspace.py         # 工作区
+│   ├── base/                # 核心类、Hooks
+│   │   ├── evo_agent.py     # 基于 AgentScope 的核心实现
+│   │   └── hooks.py         # 生命周期 Hooks
+│   └── prompts/             # Agent 提示词
+│       └── analyst/personas.yaml
+│
+├── apps/                  # 微服务入口
+│   ├── runtime_service.py   # 运行时服务（端口 8003）
+│   ├── agent_service.py    # Agent 服务（端口 8000）
+│   ├── trading_service.py  # 交易服务（端口 8001）
+│   ├── news_service.py     # 新闻服务（端口 8002）
+│   └── cors.py
+│
+├── runtime/               # 运行时管理层
+│   ├── manager.py          # TradingRuntimeManager
+│   ├── agent_runtime.py    # AgentRuntimeState
+│   ├── context.py          # TradingRunContext
+│   ├── session.py          # TradingSessionKey
+│   └── registry.py         # RuntimeRegistry
+│
+├── process/               # 进程监管层
+│   ├── supervisor.py       # ProcessSupervisor
+│   ├── registry.py         # RunRegistry
+│   └── models.py           # ProcessRun、ProcessRunState
+│
+├── core/                  # Pipeline 执行
+│   ├── pipeline.py         # TradingPipeline（核心编排器）
+│   ├── pipeline_runner.py  # 独立 Pipeline 执行
+│   ├── scheduler.py        # 调度器
+│   └── state_sync.py       # 状态同步
+│
+├── services/              # Gateway 和服务
+│   ├── gateway.py          # WebSocket 网关
+│   ├── gateway_*.py       # Gateway 子模块
+│   ├── market.py           # 市场数据服务
+│   ├── storage.py          # 存储服务
+│   ├── runtime_db.py       # 运行时数据库
+│   └── research_db.py      # 研究数据库
+│
+├── data/                  # 市场数据处理
+│   ├── provider_router.py   # 数据源路由
+│   ├── provider_utils.py   # 数据源工具
+│   ├── market_store.py     # 市场数据存储
+│   ├── market_ingest.py    # 数据采集
+│   ├── cache.py            # 缓存
+│   ├── schema.py           # 数据 schema
+│   ├── historical_price_manager.py  # 历史价格管理
+│   ├── polling_price_manager.py     # 轮询价格管理
+│   ├── news_alignment.py   # 新闻对齐
+│   ├── polygon_client.py   # Polygon.io 客户端
+│   └── ret_data_updater.py # 离线数据更新
+│
+├── config/                # 配置
+│   ├── constants.py        # Agent 配置、显示名称
+│   ├── bootstrap_config.py # 启动配置解析
+│   ├── env_config.py      # 环境变量配置
+│   ├── data_config.py     # 数据源配置
+│   └── agent_profiles.yaml # Agent Profile 配置
+│
 ├── domains/               # 领域业务逻辑
 │   ├── news.py
 │   └── trading.py
-├── services/         # Gateway 和辅助服务
+│
-│   ├── gateway.py           # 统一路由网关
-│   ├── gateway_*.py         # Gateway 子模块
-│   └── market.py            # 市场数据服务
-├── api/              # FastAPI 端点
-├── config/           # 常量和配置
-│   └── constants.py   # Agent 配置、显示名称等
-├── core/             # Pipeline 执行逻辑
-├── data/             # 市场数据处理
-│   ├── provider_router.py  # 数据源路由
-│   └── schema.py          # 数据 schema
-├── enrich/           # LLM 响应富化
-├── explain/          # 交易决策解释
 ├── llm/                   # LLM 集成
 │   └── models.py          # RetryChatModel、TokenRecordingModelWrapper
-├── skills/           # 技能定义（内置 + 自定义）
+│
+├── skills/                 # 技能定义
 ├── tools/                  # 交易和分析工具
-└── utils/            # 工具函数
+├── enrich/                 # LLM 响应富化
+├── explain/                # 交易决策解释
+├── utils/                  # 工具函数
+│   ├── settlement.py       # 结算协调器
+│   ├── trade_executor.py   # 交易执行器
+│   ├── terminal_dashboard.py  # 终端仪表板
+│   ├── analyst_tracker.py  # 分析师追踪
+│   ├── baselines.py        # 基准线
+│   ├── msg_adapter.py      # 消息适配器
+│   └── progress.py         # 进度追踪
+│
+├── api/                    # FastAPI 端点
+│   └── runtime.py
+│
+└── main.py                 # 主入口点
 ```
 
 ## 前端结构
 
 ```
 frontend/src/
-├── App.jsx              # React 主应用
+├── App.jsx                 # 主应用（LiveTradingApp）
-├── components/          # React 组件
+├── AppShell.jsx            # App 外壳（布局、侧边栏）
+├── components/
 │   ├── RuntimeView.jsx          # 交易运行时 UI
 │   ├── TraderView.jsx           # 交易员界面
 │   ├── RoomView.jsx             # 聊天室视图
 │   ├── StockExplainView.jsx     # 股票解释视图
 │   ├── RuntimeSettingsPanel.jsx  # 运行时设置面板
+│   ├── RuntimeLogsModal.jsx      # 运行时日志弹窗
 │   ├── WatchlistPanel.jsx       # 关注列表
 │   ├── PerformanceView.jsx      # 绩效视图
 │   ├── StatisticsView.jsx       # 统计视图
 │   ├── NetValueChart.jsx        # 净值曲线图
 │   ├── AgentCard.jsx            # Agent 卡片
 │   ├── AgentFeed.jsx            # Agent 动态
-│   └── explain/                  # 解释相关组件
+│   ├── Header.jsx               # 头部
+│   ├── MarkdownModal.jsx        # Markdown 弹窗
+│   ├── StockLogo.jsx            # 股票 Logo
+│   └── explain/                 # 解释组件
 │       ├── ExplainNewsSection.jsx
 │       ├── ExplainRangeSection.jsx
 │       ├── ExplainSimilarDaysSection.jsx
 │       ├── ExplainStorySection.jsx
 │       └── useExplainModel.js
-├── services/            # API 服务
+├── hooks/                     # React Hooks
-│   ├── runtimeApi.js     # 运行时 API 调用
+│   ├── useWebSocketConnection.js   # WebSocket 连接管理
-│   ├── websocket.js      # WebSocket 实时通信
+│   ├── useRuntimeControls.js       # 运行时配置管理
-│   ├── newsApi.js        # 新闻服务客户端
+│   ├── useAgentDataRequests.js     # Agent 数据请求
-│   └── tradingApi.js     # 交易服务客户端
+│   ├── useStockDataRequests.js     # 股票数据请求
-├── config/
+│   ├── useStockExplainData.js      # 股票解释数据
-│   └── constants.js      # Agent 定义、配置
+│   ├── useAgentWorkspacePanel.js   # Agent 工作区面板
-└── hooks/               # React Hooks
+│   ├── useWebsocketSessionSync.js  # WebSocket 会话同步
+│   └── useFeedProcessor.js        # Feed 事件处理
+├── store/                      # Zustand 状态管理
+│   ├── runtimeStore.js         # 连接状态、运行时配置
+│   ├── marketStore.js          # 市场数据、股票价格
+│   ├── portfolioStore.js      # 组合、持仓、交易
+│   ├── agentStore.js          # Agent 技能、工作区
+│   └── uiStore.js            # UI 状态、视图切换
+├── services/
+│   ├── websocket.js           # WebSocket 客户端
+│   ├── runtimeApi.js          # 运行时 API
+│   ├── runtimeControls.js     # 运行时控制
+│   ├── newsApi.js             # 新闻 API
+│   └── tradingApi.js          # 交易 API
+├── utils/
+│   ├── formatters.js          # 格式化工具
+│   └── modelIcons.js          # 模型图标
+└── config/
+    └── constants.js           # Agent 定义、配置
 ```
 
 ## Agent 系统
@@ -195,108 +292,85 @@ frontend/src/
 | `sentiment_analyst` | 情绪分析师 | 市场情绪、新闻情绪、内幕交易 |
 | `valuation_analyst` | 估值分析师 | DCF、EV/EBITDA、intrinsic value |
 | `portfolio_manager` | 投资经理 | 决策执行、交易协调 |
-| `risk_manager` | 风控经理 | 实时价格/波动率监控、仓位限制、多层风险预警 |
+| `risk_manager` | 风控经理 | 实时价格/波动率监控、仓位限制 |
-
-### Hook 系统 (`base/hooks.py`)
-
-- **MemoryCompactionHook**: 基于 CoPaw 的内存压缩
-  - `memory_compact_ratio`: 压缩目标比例（默认 0.75）
-  - `memory_reserve_ratio`: 保留比例（默认 0.1）
-  - `enable_tool_result_compact`: 工具结果压缩
-  - `tool_result_compact_keep_n`: 保留最近 N 条工具结果
 
 ### 添加自定义分析师
 
-1. 在 `backend/agents/prompts/analyst/personas.yaml` 注册
+1. `backend/agents/prompts/analyst/personas.yaml` 注册
-2. 在 `backend/config/constants.py` 的 `ANALYST_TYPES` 字典中添加
+2. `backend/config/constants.py` 的 `ANALYST_TYPES` 字典添加
-3. 可选：在 `frontend/src/config/constants.js` 中更新前端配置
+3. `frontend/src/config/constants.js` 可选更新
 
 ### LLM 模型封装 (`backend/llm/models.py`)
 
-基于 CoPaw 的模型封装设计：
+- **RetryChatModel**: 自动重试瞬态 LLM 错误，指数退避
-
+- **TokenRecordingModelWrapper**: 追踪 token 消耗和成本
-- **RetryChatModel**: 自动重试瞬态 LLM 错误（rate limit、timeout、502/503 等），指数退避
-  - `max_retries`: 最大重试次数（默认 3）
-  - `initial_delay`: 初始延迟秒数（默认 1.0）
-  - `backoff_multiplier`: 退避倍数（默认 2.0）
-
-- **TokenRecordingModelWrapper**: 追踪每个 provider 的 token 消耗和成本
-
-```python
-from backend.llm.models import create_model, RetryChatModel
-
-model = RetryChatModel(create_model("gpt-4o", "OPENAI"), max_retries=3)
-```
 
 ## 技能系统 (`backend/skills/`)
 
-技能定义在 `SKILL.md` 文件中，包含：
+技能定义在 `SKILL.md`，包含 `instructions`、`triggers`、`parameters`、`available_tools`。
-- `instructions` - 技能说明
-- `triggers` - 触发条件
-- `parameters` - 输入/输出 schema
-- `available_tools` - 技能可使用的工具
-
-技能由 `skills_manager.py` 加载，通过 `skill_adaptation_hook.py` 绑定到 Agent。
 
 技能管理器支持 6 种作用域：builtin、customized、installed、active、disabled、local。
 
-## Pipeline 执行 (`backend/core/`)
+## 运行时数据布局
 
-每日交易流程：
+- `data/market_research.db` - 持久研究数据
-
+- `runs/<run_id>/` - 每次任务运行的状态
-1. **分析阶段** - 各 Agent 基于工具和历史经验独立分析
+- `runs/<run_id>/team_dashboard/*.json` - 仪表板导出层（非权威源）
-2. **沟通阶段** - 通过私聊、通知、会议等方式交换观点（1v1/1vN/NvN）
+- `runs/<run_id>/state/runtime_state.json` - 运行时快照
-3. **决策阶段** - 投资经理综合判断，给出最终交易
+- 运行时 API 优先使用 `server_state.json` 和 `runtime.db`
-4. **评估阶段** - 绩效跟踪
-5. **复盘阶段** - Agent 根据当日实际收益反思总结，通过 ReMe 记忆框架更新经验
-
-## 前端状态管理
-
-项目正在向 Zustand 状态管理过渡，已创建的 store：
 
 ```bash
-frontend/src/store/
+RUNS_RETENTION_COUNT=20  # 时间戳格式文件夹自动清理
-├── index.js           # 导出所有 store
-├── runtimeStore.js    # 连接状态、运行时配置
-├── marketStore.js     # 市场数据、股票价格
-├── portfolioStore.js  # 组合、持仓、交易
-├── agentStore.js      # Agent 技能、工作区
-└── uiStore.js         # UI 状态、视图切换
 ```
 
-**迁移状态**：
-- Stores 已创建但尚未在 App.jsx 中使用
-- 计划：逐步迁移 60+ 个 useState 到对应 store
-
 ## 环境配置
 
-`.env` 必需配置：
+### Backend (`env.template`)
 
 ```bash
-# 金融数据源
+# 金融数据源（支持多源fallback）
-FIN_DATA_SOURCE=finnhub|financial_datasets
+FIN_DATA_SOURCE=finnhub|financial_datasets|yfinance|local_csv
+ENABLED_DATA_SOURCES=financial_datasets,finnhub,yfinance,local_csv
 FINANCIAL_DATASETS_API_KEY=   # 回测必需
 FINNHUB_API_KEY=             # 实盘必需
+POLYGON_API_KEY=             # Polygon市场库采集可选
 
-# Agent LLM
+# LLM 配置
 OPENAI_API_KEY=
 OPENAI_BASE_URL=
 MODEL_NAME=qwen3-max-preview
 
-# 可为不同 Agent 指定不同模型
+# Agent 特定模型
-AGENT_SENTIMENT_ANALYST_MODEL_NAME=qwen3-max-preview
+AGENT_SENTIMENT_ANALYST_MODEL_NAME=deepseek-v3.2-exp
-AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=deepseek-chat
+AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4.6
+AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=qwen3-max-preview
+AGENT_VALUATION_ANALYST_MODEL_NAME=Moonshot-Kimi-K2-Instruct
+AGENT_RISK_MANAGER_MODEL_NAME=qwen3-max-preview
+AGENT_PORTFOLIO_MANAGER_MODEL_NAME=qwen3-max-preview
 
 # ReMe 记忆系统
 MEMORY_API_KEY=
+MEMORY_MODEL_NAME=qwen3-max
+MEMORY_EMBEDDING_MODEL=text-embedding-v4
+
+# 交易参数
+MAX_COMM_CYCLES=2
+MARGIN_REQUIREMENT=0.5
+DATA_START_DATE=2022-01-01
+AUTO_UPDATE_DATA=true
+```
+
+### Frontend (`frontend/env.template`)
+
+```bash
+VITE_WS_URL=ws://localhost:8765
 ```
 
 ## 关键依赖
 
 - **AgentScope** - 多智能体框架
 - **ReMe** - 持续学习记忆系统
-- **FastAPI** + **uvicorn** - 后端 API 服务器
+- **FastAPI** + **uvicorn** - 后端 API
 - **websockets** - 实时通信
 - **React 19** + **Vite** + **TailwindCSS** - 前端
-- **React Context** - 前端状态管理（App.jsx 中使用 useState + useCallback）
+- **Zustand** - 状态管理
-- **Three.js** / **React-Three-Fiber** - 3D 可视化

README.md

@@ -1,36 +1,34 @@
 <p align="center">
-  <img src="./docs/assets/evotraders_logo.jpg" width="45%">
+  <img src="./docs/assets/bigtime_logo.jpg" width="45%">
 </p>
 
-<h2 align="center">EvoTraders: A Self-Evolving Multi-Agent Trading System</h2>
+<h2 align="center">大时代：自进化多智能体交易系统</h2>
 
 <p align="center">
-  📌 <a href="http://trading.evoagents.cn">Visit us at EvoTraders website !</a>
+  📌 <a href="http://trading.evoagents.cn">Visit the 大时代 website</a>
 </p>
 
-![System Demo](./docs/assets/evotraders_demo.gif)
+![System Demo](./docs/assets/bigtime_demo.gif)
 
-EvoTraders is an open-source financial trading agent framework that builds a trading system capable of continuous learning and evolution in real markets through multi-agent collaboration and memory systems.
+大时代 is an open-source financial trading agent framework that combines multi-agent collaboration, run-scoped workspaces, and memory to support both backtests and live trading workflows.
+
+The repository name and CLI entrypoints still use `evotraders` for compatibility, but the product-facing branding now follows the 大时代 naming used by the reference branch.
 
 ---
 
 ## Core Features
 
-**Multi-Agent Collaborative Trading**
+**Multi-agent trading team**
-A team of 6 members, including 4 specialized analyst roles (fundamentals, technical, sentiment, valuation) + portfolio manager + risk management, collaborating to make decisions like a real trading team.
+Six roles collaborate like a real desk: four specialist analysts (fundamentals, technical, sentiment, valuation), one portfolio manager, and one risk manager.
 
-You can customize your Agents here: [Custom Configuration](#custom-configuration)
+**Continuous learning**
+Agents can persist long-term memory with ReMe, reflect after each cycle, and evolve their decision patterns over time.
 
-**Continuous Learning and Evolution**
+**Backtest and live modes**
-Based on the ReMe memory framework, agents reflect and summarize after each trade, preserving experience across rounds, and forming unique investment methodologies.
+The same runtime model supports historical simulation and live execution with real-time market data.
 
-Through this design, we hope that when AI Agents form a team and enter the real-time market, they will gradually develop their own trading styles and decision preferences, rather than one-time random inference.
+**Operator-facing UI**
-
+The frontend exposes the trading room, runtime controls, logs, approvals, agent workspaces, and explain/news views.
-**Real-Time Market Trading**
-Supports real-time market data integration, providing backtesting mode and live trading mode, allowing AI Agents to learn and make decisions in real market fluctuations.
-
-**Visualized Trading Information**
-Observe agents' analysis processes, communication records, and decision evolution in real-time, with complete tracking of return curves and analyst performance.
 
 <p>
   <img src="docs/assets/performance.jpg" width="45%">
@@ -39,198 +37,325 @@ Observe agents' analysis processes, communication records, and decision evolutio
 
 ---
 
+## Current Architecture
+
+The repository is currently in a transition from a modular monolith to split service surfaces. The split-service path is the default local development mode.
+
+Current app surfaces:
+
+- `backend.apps.agent_service` on `:8000`: control plane for workspaces, agents, skills, and guard/approval APIs
+- `backend.apps.trading_service` on `:8001`: read-only trading data APIs
+- `backend.apps.news_service` on `:8002`: read-only explain/news APIs
+- `backend.apps.runtime_service` on `:8003`: runtime lifecycle APIs
+- `backend.apps.openclaw_service` on `:8004`: read-only OpenClaw facade
+- WebSocket gateway on `:8765`: live event/feed channel for the frontend
+
+The most important runtime path today is:
+
+`frontend -> runtime_service/control APIs -> gateway/runtime manager -> market service + pipeline + storage`
+
+Reference notes for the migration live in [services/README.md](./services/README.md).
+
+---
+
 ## Quick Start
 
-### Installation
+### 1. Install
 
 ```bash
-# Clone repository
+# clone this repository, then:
-git clone https://github.com/agentscope-ai/agentscope-samples
+cd evotraders
-cd agentscope-samples/EvoTraders
 
-# Install dependencies (Recommend uv!)
+# backend runtime dependencies
+uv pip install -r requirements.txt
+
+# install package entrypoint in editable mode
 uv pip install -e .
-# optional: pip install -e .
 
+# optional
+# uv pip install -e ".[dev]"
+# pip install -e .
+```
 
-# Configure environment variables
+Frontend dependencies:
+
+```bash
+cd frontend
+npm ci
+cd ..
+```
+
+Production deployment should prefer `requirements.txt` for backend and `npm ci` for frontend so the pulled environment matches the checked-in lockfiles and version pins.
+
+### 2. Configure environment
+
+```bash
 cp env.template .env
-# Edit .env file and add your API Keys. The following config are required:
+```
 
-# finance data API: At minimum, FINANCIAL_DATASETS_API_KEY is required, corresponding to FIN_DATA_SOURCE=financial_datasets; It is recommended to add FINNHUB_API_KEY, corresponding to FIN_DATA_SOURCE=finnhub; If using live mode, FINNHUB_API_KEY must be added
+The root `env.template` is the canonical local template. A `.env.example` is also kept in the repo for reference.
-FIN_DATA_SOURCE =   #finnhub or financial_datasets
-FINANCIAL_DATASETS_API_KEY=  #Required
-FINNHUB_API_KEY=  #Optional
 
-# LLM API for Agents
+Minimum useful variables:
+
+```bash
+# watchlist
+TICKERS=AAPL,MSFT,GOOGL,NVDA,TSLA,META,AMZN
+
+# market data
+FIN_DATA_SOURCE=finnhub
+FINANCIAL_DATASETS_API_KEY=
+FINNHUB_API_KEY=
+POLYGON_API_KEY=
+
+# agent model
 OPENAI_API_KEY=
 OPENAI_BASE_URL=
 MODEL_NAME=qwen3-max-preview
 
-# LLM & embedding API for Memory
+# memory (optional unless --enable-memory is used)
 MEMORY_API_KEY=
 ```
 
-### Running
+Notes:
+
+- `FINNHUB_API_KEY` is required for live mode.
+- `POLYGON_API_KEY` enables long-lived market-store ingestion and refresh helpers.
+- `MEMORY_API_KEY` is only required when long-term memory is enabled.
+
+For a production-style local start flow, you can also use:
 
-**Backtest Mode:**
 ```bash
-evotraders backtest --start 2025-11-01 --end 2025-12-01
+./start.sh
-evotraders backtest --start 2025-11-01 --end 2025-12-01 --enable-memory # Use Memory
 ```
 
-If you do not have market data APIs and just want to try the backtest demo, download the offline data and unzip it into `backend/data`:
+### 3. Start the stack
+
+Recommended local development flow:
+
+```bash
+./start-dev.sh
+```
+
+This starts:
+
+- `agent_service` at `http://localhost:8000`
+- `trading_service` at `http://localhost:8001`
+- `news_service` at `http://localhost:8002`
+- `runtime_service` at `http://localhost:8003`
+- gateway WebSocket at `ws://localhost:8765`
+
+Then start the frontend in another terminal:
+
+```bash
+evotraders frontend
+```
+
+Open `http://localhost:5173`.
+
+You can also run services manually:
+
+```bash
+python -m uvicorn backend.apps.agent_service:app --host 0.0.0.0 --port 8000 --reload
+python -m uvicorn backend.apps.trading_service:app --host 0.0.0.0 --port 8001 --reload
+python -m uvicorn backend.apps.news_service:app --host 0.0.0.0 --port 8002 --reload
+python -m uvicorn backend.apps.runtime_service:app --host 0.0.0.0 --port 8003 --reload
+python -m backend.main --mode live --host 0.0.0.0 --port 8765
+```
+
+### 4. Run backtest or live mode from CLI
+
+Backtest:
+
+```bash
+evotraders backtest --start 2025-11-01 --end 2025-12-01
+evotraders backtest --start 2025-11-01 --end 2025-12-01 --enable-memory
+evotraders backtest --config-name smoke_fullstack --start 2025-11-01 --end 2025-12-01
+```
+
+Live:
+
+```bash
+evotraders live
+evotraders live --enable-memory
+evotraders live --schedule-mode intraday --interval-minutes 60
+evotraders live --trigger-time 22:30
+```
+
+Help:
+
+```bash
+evotraders --help
+evotraders backtest --help
+evotraders live --help
+evotraders frontend --help
+```
+
+### Offline backtest data
+
+If you want a quick backtest demo without external market APIs, download the offline bundle and unzip it into `backend/data`:
+
 ```bash
 wget "https://agentscope-open.oss-cn-beijing.aliyuncs.com/ret_data.zip"
 unzip ret_data.zip -d backend/data
 ```
-The zip includes basic stock price data so you can run the backtest demo out of the box.
-
-**Live Trading:**
-```bash
-evotraders live                    # Run immediately (default)
-evotraders live --enable-memory    # Use memory
-evotraders live --mock             # Mock mode (testing)
-evotraders live -t 22:30           # Run daily at 22:30 local time (auto-converts to NYSE timezone)
-```
-
-**Get Help:**
-```bash
-evotraders --help           # View global CLI help
-evotraders backtest --help  # View backtest mode parameters
-evotraders live --help      # View live/mock run parameters
-```
-
-**Launch Visualization Interface:**
-```bash
-# Ensure npm is installed, otherwise install it:
-# npm install
-evotraders frontend                # Default connects to port 8765, you can modify the address in ./frontend/env.local to change the port number
-```
-
-Visit `http://localhost:5173/` to view the trading room, select a date and click Run/Replay to observe the decision-making process.
 
 ---
 
-## System Architecture
+## Runtime Data Layout
 
-![Architecture Diagram](docs/assets/evotraders_pipeline.jpg)
+- Long-lived research data lives in `data/market_research.db`
+- Each run writes run-scoped state under `runs/<run_id>/`
+- `runs/<run_id>/BOOTSTRAP.md` stores run-specific bootstrap values and prompt body
+- `runs/<run_id>/state/runtime_state.json` stores runtime snapshot state
+- `runs/<run_id>/team_dashboard/*.json` is a compatibility/export layer for dashboard consumers, not the primary runtime source of truth
 
-### Agent Design
+Optional retention control:
 
-**Analyst Team:**
+```bash
-- **Fundamentals Analyst**: Financial health, profitability, growth quality
+RUNS_RETENTION_COUNT=20
-- **Technical Analyst**: Price trends, technical indicators, momentum analysis
-- **Sentiment Analyst**: Market sentiment, news sentiment, insider trading
-- **Valuation Analyst**: DCF, residual income, EV/EBITDA
-
-**Decision Layer:**
-- **Portfolio Manager**: Integrates analysis signals from analysts, executes communication strategies, combines analyst and team historical performance, recent investment memories, and long-term investment experience to make final decisions
-- **Risk Management**: Real-time price and volatility monitoring, position limits, multi-layer risk warnings
-
-### Decision Process
-
-```
-Real-time Market Data → Independent Analysis → Intelligent Communication (1v1/1vN/NvN) → Decision Execution → Performance Evaluation → Learning and Evolution (Memory Update)
 ```
 
-Each trading day goes through five stages:
+Only timestamped run folders like `YYYYMMDD_HHMMSS` are pruned automatically. Named runs such as `live`, `smoke_fullstack`, or `reload_demo_*` are preserved.
-
-1. **Analysis Stage**: Each agent independently analyzes based on their respective tools and historical experience
-2. **Communication Stage**: Exchange views through private chats, notifications, meetings, etc.
-3. **Decision Stage**: Portfolio manager makes comprehensive judgments and provides final trades
-4. **Evaluation Stage**
-   - **Performance Charts**: Track portfolio return curves vs. benchmark strategies (equal-weighted, market-cap weighted, momentum). Used to evaluate overall strategy effectiveness.
-
-   - **Analyst Rankings**: Click on avatars in the Trading Room to view analyst performance (win rate, bull/bear market win rate). Used to understand which analysts provide the most valuable insights.
-
-   - **Statistics**: Detailed position and trading history. Used for in-depth analysis of position management and execution quality.
-
-5. **Review Stage**: Agents reflect on decisions and summarize experiences based on actual returns of the day, and store them in the ReMe memory framework for continuous improvement
 
 ---
 
-### Module Support
+## Frontend Service Routing
 
-- **Agent Framework**: [AgentScope](https://github.com/agentscope-ai/agentscope)
+The frontend always uses the control plane and runtime APIs, and can optionally call split services directly for read-only data.
-- **Memory System**: [ReMe](https://github.com/agentscope-ai/reme)
+
-- **LLM Support**: OpenAI, DeepSeek, Qwen, Moonshot, Zhipu AI, etc.
+Useful frontend env vars:
+
+```bash
+VITE_CONTROL_API_BASE_URL=http://localhost:8000/api
+VITE_RUNTIME_API_BASE_URL=http://localhost:8003/api/runtime
+VITE_NEWS_SERVICE_URL=http://localhost:8002
+VITE_TRADING_SERVICE_URL=http://localhost:8001
+VITE_WS_URL=ws://localhost:8765
+```
+
+If these are not set, the frontend falls back to its local defaults and compatibility paths where available.
+
+---
+
+## Decision Flow
+
+```text
+Market data -> independent analyst work -> team communication -> portfolio decision ->
+risk review -> execution/settlement -> reflection/memory update
+```
+
+The runtime manager also tracks:
+
+- agent registration and status
+- pending approvals
+- run events
+- current session key
 
 ---
 
 ## Custom Configuration
 
-### Custom Analyst Roles
+### Add or change analyst roles
 
-1. Register role information in [./backend/agents/prompts/analyst/personas.yaml](./backend/agents/prompts/analyst/personas.yaml), for example:
+1. Define the analyst persona in [backend/agents/prompts/analyst/personas.yaml](./backend/agents/prompts/analyst/personas.yaml)
+2. Register the role in [backend/config/constants.py](./backend/config/constants.py)
+3. Optionally add/update the frontend seat metadata in [frontend/src/config/constants.js](./frontend/src/config/constants.js)
+
+Example persona entry:
 
 ```yaml
 comprehensive_analyst:
   name: "Comprehensive Analyst"
   focus:
-    - ...
+    - multi-factor synthesis
-  preferred_tools: # Flexibly select based on situation
+  preferred_tools:
+    - get_stock_price
+    - get_company_financials
   description: |
-    As a comprehensive analyst ...
+    A generalist analyst that combines multiple signals.
 ```
 
-2. Add role definition in [./backend/config/constants.py](./backend/config/constants.py)
+### Configure per-agent models
-```python
-ANALYST_TYPES = {
-    # Add new analyst
-    "comprehensive_analyst": {
-        "display_name": "Comprehensive Analyst",
-        "agent_id": "comprehensive_analyst",
-        "description": "Uses LLM to intelligently select analysis tools, performs comprehensive analysis",
-        "order": 15
-    }
-}
-```
 
-3. Introduce new role in frontend configuration [./frontend/src/config/constants.js](./frontend/src/config/constants.js) (optional)
+Model overrides are configured in `.env`:
-```javascript
-export const AGENTS = [
-    // Override one of the agents
-  {
-    id: "comprehensive_analyst",
-    name: "Comprehensive Analyst",
-    role: "Comprehensive Analyst",
-    avatar: `${ASSET_BASE_URL}/...`,
-    colors: { bg: '#F9FDFF', text: '#1565C0', accent: '#1565C0' }
-  }
-]
-```
-
-### Custom Models
-
-Configure models used by different agents in the [.env](.env) file:
 
 ```bash
-AGENT_SENTIMENT_ANALYST_MODEL_NAME=qwen3-max-preview
+AGENT_SENTIMENT_ANALYST_MODEL_NAME=deepseek-v3.2-exp
-AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=deepseek-chat
+AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4.6
-AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4-plus
+AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=qwen3-max-preview
-AGENT_VALUATION_ANALYST_MODEL_NAME=moonshot-v1-32k
+AGENT_VALUATION_ANALYST_MODEL_NAME=Moonshot-Kimi-K2-Instruct
+AGENT_RISK_MANAGER_MODEL_NAME=qwen3-max-preview
+AGENT_PORTFOLIO_MANAGER_MODEL_NAME=qwen3-max-preview
 ```
 
-### Project Structure
+### Run-scoped bootstrap config
 
+Each run can override defaults through `runs/<run_id>/BOOTSTRAP.md`. The front matter is parsed by [backend/config/bootstrap_config.py](./backend/config/bootstrap_config.py) and can define values such as:
+
+```yaml
+tickers:
+  - AAPL
+  - MSFT
+initial_cash: 100000
+margin_requirement: 0.5
+max_comm_cycles: 2
+schedule_mode: daily
+trigger_time: "09:30"
+enable_memory: false
 ```
-EvoTraders/
+
+Initialize a run workspace with:
+
+```bash
+evotraders init-workspace --config-name my_run
+```
+
+---
+
+## Project Structure
+
+```text
+evotraders/
 ├── backend/
-│   ├── agents/           # Agent implementation
+│   ├── agents/      # agent roles, prompts, skills, workspaces
-│   ├── communication/    # Communication system
+│   ├── api/         # FastAPI routers
-│   ├── memory/          # Memory system (ReMe)
+│   ├── apps/        # split service surfaces
-│   ├── tools/           # Analysis toolset
+│   ├── core/        # pipeline, scheduler, state sync
-│   ├── servers/         # WebSocket services
+│   ├── runtime/     # runtime manager and agent runtime state
-│   └── cli.py           # CLI entry point
+│   ├── services/    # gateway, market/storage/db services
-├── frontend/            # React visualization interface
+│   └── cli.py       # Typer CLI entrypoint
-└── logs_and_memory/     # Logs and memory data
+├── frontend/        # React + Vite UI
+├── shared/          # shared clients and schemas for split services
+├── runs/            # run-scoped state and dashboards
+├── data/            # long-lived research artifacts
+└── services/README.md
+```
+
+---
+
+## Testing
+
+Backend tests live under `backend/tests` and cover service apps, shared clients, domains, routing, enrichment, gateway support, and runtime support.
+
+Typical commands:
+
+```bash
+pytest
+pytest backend/tests/test_runtime_service_app.py
+pytest backend/tests/test_trading_service_app.py
+```
+
+Frontend tests:
+
+```bash
+cd frontend
+npm test
 ```
 
 ---
 
 ## License and Disclaimer
 
-EvoTraders is a research and educational project, open-sourced under the Apache 2.0 license.
+大时代 is a research and educational project. Review the repository license before redistribution or commercial use.
 
-**Risk Warning**: Before trading with real funds, please conduct thorough testing and risk assessment. Past performance does not guarantee future returns. Investment involves risks, and decisions should be made with caution.
+**Risk warning**: this project is not investment advice. Test thoroughly before any real-money deployment. Past performance does not guarantee future returns.

README_zh.md

@@ -1,294 +1,359 @@
 <p align="center">
-  <img src="./docs/assets/evotraders_logo.jpg" width="45%">
+  <img src="./docs/assets/bigtime_logo.jpg" width="45%">
 </p>
 
-<h2 align="center">EvoTraders：自我进化的多智能体交易系统</h2>
+<h2 align="center">大时代：自进化多智能体交易系统</h2>
-
 
 <p align="center">
-  📌 <a href="http://trading.evoagents.cn">Visit us at EvoTraders website !</a>
+  📌 <a href="http://trading.evoagents.cn">访问大时代官网</a>
 </p>
 
-![系统演示](./docs/assets/evotraders_demo.gif)
+![系统演示](./docs/assets/bigtime_demo.gif)
 
-EvoTraders是一个开源的金融交易智能体框架，通过多智能体协作和记忆系统，构建能够在真实市场中持续学习与进化的交易系统。
+大时代 是一个开源的金融交易智能体框架，结合多智能体协作、run 级工作区和记忆机制，支持回测与实盘两类交易运行模式。
 
 ---
 
 ## 核心特性
 
-**多智能体协作交易**
+**多智能体交易团队**
-6名成员，包含4种专业分析师角色（基本面、技术面、情绪、估值）+ 投资组合经理 + 风险管理，像真实交易团队一样协作决策。
+系统默认包含 6 个角色：4 个分析师（基本面、技术面、情绪、估值）+ 投资经理 + 风控经理。
 
-你可以在这里自定义你的Agents，支持配置不同大模型（如 Qwen、DeepSeek、GPT、Claude等）协同分析：[自定义配置](#自定义配置)
+**持续学习**
+可选接入 ReMe 长期记忆，智能体会在每轮结束后反思、复盘并沉淀经验。
 
-**持续学习与进化**
+**统一运行时**
-基于 ReMe 记忆框架，智能体在每次交易后反思总结，跨回合保留经验，形成独特的投资方法论。
+同一套运行时模型支持历史回测和实时行情驱动的实盘流程。
-
-通过这样的设计，我们希望当 AI Agents 组成团队进入实时市场，它们会逐渐形成自己的交易风格和决策偏好，而不是一次性的随机推理
-
-
-**实时市场交易**
-支持实时行情接入，提供回测模式和实盘模式，让 AI Agents 在真实市场波动中学习和决策。
-
-**可视化交易信息**
-实时观察 Agents 的分析过程、沟通记录和决策演化，完整追踪收益曲线和分析师表现。
 
+**可操作前端**
+前端不只是展示层，还包含交易室、运行控制、日志、审批、Agent 工作区和 explain/news 视图。
 
 <p>
   <img src="docs/assets/performance.jpg" width="45%">
   <img src="./docs/assets/dashboard.jpg" width="45%">
 </p>
 
+---
+
+## 当前架构
+
+仓库目前处于“模块化单体 -> 拆分服务”的迁移阶段，本地开发默认走 split-service 路径。
+
+当前 app surface：
+
+- `backend.apps.agent_service`，端口 `8000`：控制面，负责 workspaces、agents、skills、审批接口
+- `backend.apps.trading_service`，端口 `8001`：只读交易数据接口
+- `backend.apps.news_service`，端口 `8002`：只读 explain/news 接口
+- `backend.apps.runtime_service`，端口 `8003`：运行时生命周期接口
+- `backend.apps.openclaw_service`，端口 `8004`：只读 OpenClaw facade
+- WebSocket gateway，端口 `8765`：前端实时事件和 feed 通道
+
+当前最关键的主链路是：
+
+`frontend -> runtime_service/control APIs -> gateway/runtime manager -> market service + pipeline + storage`
+
+迁移背景可参考 [services/README.md](./services/README.md)。
 
 ---
 
 ## 快速开始
 
-### 安装
+### 1. 安装
 
 ```bash
-# 克隆仓库
+# 克隆仓库后进入项目目录
-git clone https://github.com/agentscope-ai/agentscope-samples
+cd evotraders
-cd agentscope-samples/EvoTraders
 
-# 安装依赖(推荐使用uv）
+# 安装后端运行时依赖
+uv pip install -r requirements.txt
+
+# 安装项目入口（可编辑模式）
 uv pip install -e .
-# (可选）pip install -e .
 
-# 配置环境变量
+# 可选
+# uv pip install -e ".[dev]"
+# pip install -e .
+```
+
+前端依赖：
+
+```bash
+cd frontend
+npm ci
+cd ..
+```
+
+生产环境部署建议后端使用 `requirements.txt`，前端使用 `npm ci`，这样拉起的环境会严格跟随仓库中锁定的依赖版本。
+
+### 2. 配置环境变量
+
+```bash
 cp env.template .env
-# 编辑 .env 文件，添加你的 API Keys,以下的配置项为必填项
+```
 
-# finance data API:至少需要FINANCIAL_DATASETS_API_KEY，对应FIN_DATA_SOURCE=financial_datasets；推荐添加FINNHUB_API_KEY，对应至少需要FINANCIAL_DATASETS_API_KEY，对应FIN_DATA_SOURCE填为finnhub;如果使用live 模式必须添加FINNHUB_API_KEY
+根目录 `env.template` 是当前本地开发的主模板，仓库里也保留了 `.env.example` 作为参考。
-FIN_DATA_SOURCE=   #finnhub or financial_datasets
-FINANCIAL_DATASETS_API_KEY=  #必需
-FINNHUB_API_KEY=  #可选
 
-# LLM API for Agents
+最常用的配置项：
+
+```bash
+# 自选股
+TICKERS=AAPL,MSFT,GOOGL,NVDA,TSLA,META,AMZN
+
+# 行情数据
+FIN_DATA_SOURCE=finnhub
+FINANCIAL_DATASETS_API_KEY=
+FINNHUB_API_KEY=
+POLYGON_API_KEY=
+
+# Agent 模型
 OPENAI_API_KEY=
 OPENAI_BASE_URL=
 MODEL_NAME=qwen3-max-preview
 
-# LLM & embedding API for Memory
+# 长期记忆（只有启用 --enable-memory 才需要）
 MEMORY_API_KEY=
 ```
 
-### 运行
+说明：
+
+- live 模式必须配置 `FINNHUB_API_KEY`
+- `POLYGON_API_KEY` 用于长期 market store 的补数和刷新
+- `MEMORY_API_KEY` 仅在启用长期记忆时需要
+
+如果要用更接近生产的本地启动方式，也可以直接执行：
 
-**回测模式：**
 ```bash
-evotraders backtest --start 2025-11-01 --end 2025-12-01
+./start.sh
-evotraders backtest --start 2025-11-01 --end 2025-12-01 --enable-memory # 使用记忆
-
 ```
 
-如果没有可用的行情 API，想快速体验回测 demo，可直接下载离线数据并解压到 `backend/data`：
+### 3. 启动服务栈
+
+本地开发推荐直接使用：
+
+```bash
+./start-dev.sh
+```
+
+该脚本会启动：
+
+- `agent_service`：`http://localhost:8000`
+- `trading_service`：`http://localhost:8001`
+- `news_service`：`http://localhost:8002`
+- `runtime_service`：`http://localhost:8003`
+- gateway WebSocket：`ws://localhost:8765`
+
+然后在另一个终端启动前端：
+
+```bash
+evotraders frontend
+```
+
+访问 `http://localhost:5173`。
+
+也可以手动分别启动：
+
+```bash
+python -m uvicorn backend.apps.agent_service:app --host 0.0.0.0 --port 8000 --reload
+python -m uvicorn backend.apps.trading_service:app --host 0.0.0.0 --port 8001 --reload
+python -m uvicorn backend.apps.news_service:app --host 0.0.0.0 --port 8002 --reload
+python -m uvicorn backend.apps.runtime_service:app --host 0.0.0.0 --port 8003 --reload
+python -m backend.main --mode live --host 0.0.0.0 --port 8765
+```
+
+### 4. 使用 CLI 运行回测或实盘
+
+回测：
+
+```bash
+evotraders backtest --start 2025-11-01 --end 2025-12-01
+evotraders backtest --start 2025-11-01 --end 2025-12-01 --enable-memory
+evotraders backtest --config-name smoke_fullstack --start 2025-11-01 --end 2025-12-01
+```
+
+实盘：
+
+```bash
+evotraders live
+evotraders live --enable-memory
+evotraders live --schedule-mode intraday --interval-minutes 60
+evotraders live --trigger-time 22:30
+```
+
+帮助：
+
+```bash
+evotraders --help
+evotraders backtest --help
+evotraders live --help
+evotraders frontend --help
+```
+
+### 离线回测数据
+
+如果只是想快速体验回测，不依赖外部行情 API，可以下载离线数据包并解压到 `backend/data`：
+
 ```bash
 wget "https://agentscope-open.oss-cn-beijing.aliyuncs.com/ret_data.zip"
 unzip ret_data.zip -d backend/data
 ```
-该压缩包提供基础的股票行情数据，解压后即可直接用于回测演示。
 
-**实盘交易：**
+---
+
+## 运行时数据布局
+
+- 长期研究数据保存在 `data/market_research.db`
+- 每次 run 的状态写入 `runs/<run_id>/`
+- `runs/<run_id>/BOOTSTRAP.md` 保存该 run 的 bootstrap 值和 prompt body
+- `runs/<run_id>/state/runtime_state.json` 保存运行时快照
+- `runs/<run_id>/team_dashboard/*.json` 主要是给 dashboard 用的兼容导出层，不是唯一真相源
+
+可选保留策略：
+
 ```bash
-evotraders live                    # 立即运行（默认）
+RUNS_RETENTION_COUNT=20
-evotraders live --enable-memory    # 使用记忆
-evotraders live --mock             # Mock 模式（测试）
-evotraders live -t 22:30           # 每天本地时间 22:30 运行（自动转换为 NYSE 时区）
-evotraders live --schedule-mode intraday --interval-minutes 60  # 每隔 1 小时触发一次；仅交易时段执行交易，其他时段只分析
 ```
 
-前端的“运行设置”面板也支持热更新 `schedule_mode`、`interval_minutes`、`max_comm_cycles`；其中 daily 模式时间当前按 NYSE/ET 配置。
+只有形如 `YYYYMMDD_HHMMSS` 的时间戳目录会被自动清理；`live`、`smoke_fullstack`、`reload_demo_*` 这类命名 run 会保留。
 
-**获取帮助：**
+---
-```bash
-evotraders --help           # 查看整体命令行帮助
-evotraders backtest --help  # 查看回测模式的参数说明
-evotraders live --help      # 查看实盘/Mock 运行的参数说明
-```
 
-**启动可视化界面：**
+## 前端服务路由
-```bash
-# 确保已安装 npm, 否则请安装：
-# npm install
-evotraders frontend                # 默认连接 8765 端口, 你可以修改 ./frontend/env.local 中的地址从而修改端口号
-```
 
-访问 `http://localhost:5173/` 查看交易大厅，选择日期并点击 Run/Replay 观察决策过程。
+前端始终会使用 control plane 和 runtime API，同时可以选择直连拆分服务读取只读数据。
 
-### 迁移期服务边界说明
+常用前端环境变量：
-
-当前仓库正处于从模块化单体向独立服务迁移的阶段，当前默认开发路径已经切到独立 app surface：
-
-- `backend.apps.agent_service`
-- `backend.apps.runtime_service`
-- `backend.apps.trading_service`
-- `backend.apps.news_service`
-
-当前本地开发默认推荐直接运行拆分后的服务：
 
 ```bash
-./start-dev.sh split
-
-# 或分别手动启动
-python -m uvicorn backend.apps.agent_service:app --port 8000 --reload
-python -m uvicorn backend.apps.runtime_service:app --port 8003 --reload
-python -m uvicorn backend.apps.trading_service:app --port 8001 --reload
-python -m uvicorn backend.apps.news_service:app --port 8002 --reload
-```
-
-迁移期关键环境变量：
-
-```bash
-# 后端 Gateway 优先走独立服务读取
-NEWS_SERVICE_URL=http://localhost:8002
-TRADING_SERVICE_URL=http://localhost:8001
-
-# 前端浏览器直连控制面 / 运行时面
 VITE_CONTROL_API_BASE_URL=http://localhost:8000/api
 VITE_RUNTIME_API_BASE_URL=http://localhost:8003/api/runtime
-
-# 前端浏览器优先直连独立服务
 VITE_NEWS_SERVICE_URL=http://localhost:8002
 VITE_TRADING_SERVICE_URL=http://localhost:8001
+VITE_WS_URL=ws://localhost:8765
 ```
 
-目前前端已支持直连 `news-service` 的 explain 只读路径包括：
+如果不配置，前端会按本地默认值和兼容回退逻辑运行。
-
-- runtime panel / gateway port 查询已可独立指向 `runtime-service`
-- story
-- similar days
-- range explain
-- news for date
-- news categories
-
-如果没有配置这些变量，系统会继续走当前保留的本地回退逻辑。
 
 ---
 
-## 系统架构
+## 决策流程
 
-![架构图](docs/assets/evotraders_pipeline.jpg)
+```text
-
+市场数据 -> 分析师独立分析 -> 团队沟通 -> 投资决策 ->
-### 智能体设计
+风控审核 -> 执行/结算 -> 复盘/记忆更新
-
-**分析师团队：**
-- **基本面分析师**：财务健康度、盈利能力、增长质量
-- **技术分析师**：价格趋势、技术指标、动量分析
-- **情绪分析师**：市场情绪、新闻舆情、内部人交易
-- **估值分析师**：DCF、剩余收益、EV/EBITDA
-
-**决策层：**
-- **投资组合经理**：整合来自分析师的分析信号，执行沟通策略，结合分析师和团队历史表现、近期投资记忆和长期投资经验，进行最终决策
-- **风险管理**：实时价格与波动率监控、头寸限制，多层风险预警
-
-### 决策流程
-
-```
-实时行情 → 独立分析 → 智能沟通 (1v1/1vN/NvN) → 决策执行 → 收益评估 → 学习与进化（记忆更新）
 ```
 
-每个交易日经历五个阶段：
+运行时管理器还会跟踪：
-
-1. **分析阶段**：各智能体基于各自工具和历史经验独立分析
-2. **沟通阶段**：通过私聊、通知、会议等方式交换观点
-3. **决策阶段**：投资组合经理综合判断，给出最终交易
-4. **评估阶段**
-   - **业绩图表**: 追踪组合收益曲线 vs. 基准策略（等权、市值加权、动量）。用于评估整体策略有效性。
-
-   - **分析师排名**: 在 Trading Room 点击头像查看分析师表现（胜率、牛/熊市胜率）。用于了解哪些分析师提供最有价值的洞察。
-
-   - **统计数据**: 详细的持仓和交易历史。用于深入分析仓位管理和执行质量。
-
-4. **复盘阶段**：Agents 根据当日实际收益反思决策、总结经验，并存入 ReMe 记忆框架以持续改进
-
----
-
-### 模块支持
-
-- **智能体框架**：[AgentScope](https://github.com/agentscope-ai/agentscope)
-- **记忆系统**：[ReMe](https://github.com/agentscope-ai/reme)
-- **LLM 支持**：OpenAI、DeepSeek、Qwen、Moonshot、Zhipu AI 等
 
+- agent 注册和状态
+- 待审批项
+- run 事件
+- 当前 session key
 
 ---
 
 ## 自定义配置
 
-### 自定义分析师角色
+### 新增或修改分析师角色
 
-1. 在 [./backend/agents/prompts/analyst/personas.yaml](./backend/agents/prompts/analyst/personas.yaml) 中注册角色信息，例如：
+1. 在 [backend/agents/prompts/analyst/personas.yaml](./backend/agents/prompts/analyst/personas.yaml) 中定义 persona
+2. 在 [backend/config/constants.py](./backend/config/constants.py) 中注册角色
+3. 如有需要，在 [frontend/src/config/constants.js](./frontend/src/config/constants.js) 中补充前端展示元数据
+
+示例：
 
 ```yaml
 comprehensive_analyst:
   name: "Comprehensive Analyst"
   focus:
-    - ...
+    - multi-factor synthesis
-  preferred_tools:   # Flexibly select based on situation
+  preferred_tools:
+    - get_stock_price
+    - get_company_financials
   description: |
-    As a comprehensive analyst ...
+    A generalist analyst that combines multiple signals.
 ```
 
-2. 在 [./backend/config/constants.py](./backend/config/constants.py) 添加角色定义
+### 配置各 Agent 使用的模型
-```python
-ANALYST_TYPES = {
-    # 增加新的分析师
-    "comprehensive_analyst": {
-        "display_name": "Comprehensive Analyst",
-        "agent_id": "comprehensive_analyst",
-        "description": "Uses LLM to intelligently select analysis tools, performs comprehensive analysis",
-        "order": 15
-    }
-}
-```
 
-3. 在前端配置 [./frontend/src/config/constants.js](./frontend/src/config/constants.js) 中引入新角色（可选）
+模型覆盖在 `.env` 中配置：
-```javascript
-export const AGENTS = [
-    // 覆盖掉其中某一个agent
-  {
-    id: "comprehensive_analyst",
-    name: "Comprehensive Analyst",
-    role: "Comprehensive Analyst",
-    avatar: `${ASSET_BASE_URL}/...`,
-    colors: { bg: '#F9FDFF', text: '#1565C0', accent: '#1565C0' }
-  }
-  ]
-```
-
-
-
-### 自定义模型
-
-在 [.env](.env) 文件中配置不同智能体使用的模型：
 
 ```bash
-AGENT_SENTIMENT_ANALYST_MODEL_NAME=qwen3-max-preview
+AGENT_SENTIMENT_ANALYST_MODEL_NAME=deepseek-v3.2-exp
-AGENT_FUNDAMENTAL_ANALYST_MODEL_NAME=deepseek-chat
+AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4.6
-AGENT_TECHNICAL_ANALYST_MODEL_NAME=glm-4-plus
+AGENT_FUNDAMENTALS_ANALYST_MODEL_NAME=qwen3-max-preview
-AGENT_VALUATION_ANALYST_MODEL_NAME=moonshot-v1-32k
+AGENT_VALUATION_ANALYST_MODEL_NAME=Moonshot-Kimi-K2-Instruct
+AGENT_RISK_MANAGER_MODEL_NAME=qwen3-max-preview
+AGENT_PORTFOLIO_MANAGER_MODEL_NAME=qwen3-max-preview
 ```
 
-### 项目结构
+### run 级 BOOTSTRAP 配置
 
+每个 run 都可以通过 `runs/<run_id>/BOOTSTRAP.md` 覆盖默认值。该文件由 [backend/config/bootstrap_config.py](./backend/config/bootstrap_config.py) 解析，front matter 可配置：
+
+```yaml
+tickers:
+  - AAPL
+  - MSFT
+initial_cash: 100000
+margin_requirement: 0.5
+max_comm_cycles: 2
+schedule_mode: daily
+trigger_time: "09:30"
+enable_memory: false
 ```
-EvoTraders/
+
+初始化一个 run 工作区：
+
+```bash
+evotraders init-workspace --config-name my_run
+```
+
+---
+
+## 项目结构
+
+```text
+evotraders/
 ├── backend/
-│   ├── agents/           # 智能体实现
+│   ├── agents/      # agent 角色、prompts、skills、workspaces
-│   ├── communication/    # 通信系统
+│   ├── api/         # FastAPI 路由层
-│   ├── memory/          # 记忆系统 (ReMe)
+│   ├── apps/        # 拆分服务 app surface
-│   ├── tools/           # 分析工具集
+│   ├── core/        # pipeline、scheduler、state sync
-│   ├── servers/         # WebSocket 服务
+│   ├── runtime/     # runtime manager 和 agent runtime state
-│   └── cli.py           # CLI 入口
+│   ├── services/    # gateway、market/storage/db 服务
-├── frontend/            # React 可视化界面
+│   └── cli.py       # Typer CLI 入口
-└── logs_and_memory/     # 日志和记忆数据
+├── frontend/        # React + Vite 前端
+├── shared/          # 拆分服务共用 client 和 schema
+├── runs/            # run 级状态和 dashboard 导出
+├── data/            # 长期研究数据
+└── services/README.md
+```
+
+---
+
+## 测试
+
+后端测试位于 `backend/tests`，覆盖 service app、shared client、domain、路由、enrichment、gateway 支撑模块和 runtime 支撑模块。
+
+常用命令：
+
+```bash
+pytest
+pytest backend/tests/test_runtime_service_app.py
+pytest backend/tests/test_trading_service_app.py
+```
+
+前端测试：
+
+```bash
+cd frontend
+npm test
 ```
 
 ---
 
 ## 许可与免责
 
-EvoTraders 是一个研究和教育项目，采用 Apache 2.0 许可协议开源。
+大时代 是研究和教育用途项目。再次分发或商用前，请先核对仓库中的实际 license 文件。
 
-**风险提示**：在实际资金交易前，请务必进行充分的测试和风险评估。历史表现不代表未来收益，投资有风险，决策需谨慎。
+**风险提示**：本项目不构成投资建议。任何实盘部署前都应进行充分测试和风险评估，历史表现不代表未来收益。

backend/agents/__init__.py

@@ -16,7 +16,7 @@ Exports:
 
 # New EvoAgent architecture (from agent_core.py)
 from .agent_core import EvoAgent, ToolGuardMixin, CommandHandler
-from .factory import AgentFactory, ModelConfig, RoleConfig
+from .factory import AgentFactory, ModelConfig
 from .workspace import WorkspaceManager, WorkspaceRegistry, WorkspaceConfig
 from .workspace_manager import RunWorkspaceManager
 from .registry import AgentRegistry, AgentInfo, get_registry, reset_registry
@@ -36,7 +36,6 @@ __all__ = [
     "CommandHandler",
     "AgentFactory",
     "ModelConfig",
-    "RoleConfig",
     "WorkspaceManager",
     "WorkspaceRegistry",
     "WorkspaceConfig",

backend/agents/analyst.py

@@ -84,7 +84,6 @@ class AnalystAgent(ReActAgent):
             agent_id=self.agent_id,
             config_name=self.config.get("config_name", "default"),
             toolkit=self.toolkit,
-            analyst_type=self.analyst_type_key,
         )
 
     async def reply(self, x: Msg = None) -> Msg:

backend/agents/base/__init__.py

@@ -1,5 +1,5 @@
 # -*- coding: utf-8 -*-
-"""Base agent module for EvoTraders.
+"""Base agent module for 大时代.
 
 提供Agent基础类、命令处理、工具守卫和钩子管理等功能。
 """

backend/agents/base/evo_agent.py

@@ -1,5 +1,5 @@
 # -*- coding: utf-8 -*-
-"""EvoAgent - Core agent implementation for EvoTraders.
+"""EvoAgent - Core agent implementation for 大时代.
 
 This module provides the main EvoAgent class built on AgentScope's ReActAgent,
 with integrated tools, skills, and memory management based on CoPaw design.

backend/agents/base/hooks.py

@@ -294,8 +294,8 @@ class WorkspaceWatchHook(Hook):
 
     # Files to monitor (same as PromptBuilder.DEFAULT_FILES)
     WATCHED_FILES = frozenset([
-        "SOUL.md", "AGENTS.md", "PROFILE.md", "ROLE.md",
+        "SOUL.md", "AGENTS.md", "PROFILE.md",
-        "POLICY.md", "MEMORY.md", "HEARTBEAT.md", "STYLE.md",
+        "POLICY.md", "MEMORY.md",
         "BOOTSTRAP.md",
     ])
 
@@ -601,94 +601,6 @@ class MemoryCompactionHook(Hook):
             )
 
 
-class HeartbeatHook(Hook):
-    """Pre-reasoning hook that injects HEARTBEAT.md content.
-
-    Reads the agent's HEARTBEAT.md file and prepends it to the
-    reasoning input, causing the agent to perform self-checks.
-
-    This enables "主动检查" (proactive monitoring) - periodic
-    market condition and position checks during trading hours.
-    """
-
-    HEARTBEAT_FILE = "HEARTBEAT.md"
-
-    def __init__(self, workspace_dir: Path):
-        """Initialize heartbeat hook.
-
-        Args:
-            workspace_dir: Working directory containing HEARTBEAT.md
-        """
-        self.workspace_dir = Path(workspace_dir)
-        self._completed_flag = self.workspace_dir / ".heartbeat_completed"
-
-    def _read_heartbeat_content(self) -> Optional[str]:
-        """Read HEARTBEAT.md if it exists and is non-empty.
-
-        Returns:
-            The HEARTBEAT.md content stripped of whitespace, or None
-            if the file is absent or empty.
-        """
-        hb_path = self.workspace_dir / self.HEARTBEAT_FILE
-        if not hb_path.exists():
-            return None
-        content = hb_path.read_text(encoding="utf-8").strip()
-        return content if content else None
-
-    async def __call__(
-        self,
-        agent: "ReActAgent",
-        kwargs: Dict[str, Any],
-    ) -> Optional[Dict[str, Any]]:
-        """Prepend heartbeat task to user message.
-
-        Args:
-            agent: The agent instance
-            kwargs: Input arguments to the _reasoning method
-
-        Returns:
-            Modified kwargs with heartbeat content prepended, or None
-            if no HEARTBEAT.md content is available.
-        """
-        try:
-            content = self._read_heartbeat_content()
-            if not content:
-                return None
-
-            logger.debug(
-                "Heartbeat: found HEARTBEAT.md for agent %s",
-                getattr(agent, "agent_id", "unknown"),
-            )
-
-            # Build heartbeat task instruction (Chinese)
-            hb_task = (
-                "# 定期主动检查\n\n"
-                f"{content}\n\n"
-                "请执行上述检查并报告结果。"
-            )
-
-            # Inject into the first user message in memory
-            if hasattr(agent, "memory") and agent.memory.content:
-                system_count = sum(
-                    1 for msg, _ in agent.memory.content if msg.role == "system"
-                )
-                for msg, _ in agent.memory.content[system_count:]:
-                    if msg.role == "user":
-                        original_content = msg.content
-                        msg.content = hb_task + "\n\n" + original_content
-                        break
-
-            logger.debug(
-                "Heartbeat task prepended for agent %s",
-                getattr(agent, "agent_id", "unknown"),
-            )
-
-        except Exception as e:
-            logger.error("Heartbeat hook failed: %s", e, exc_info=True)
-
-        return None
-
-
 __all__ = [
     "Hook",
     "HookManager",
@@ -696,7 +608,6 @@ __all__ = [
     "HOOK_PRE_REASONING",
     "HOOK_POST_ACTING",
     "BootstrapHook",
-    "HeartbeatHook",
     "MemoryCompactionHook",
     "WorkspaceWatchHook",
 ]

backend/agents/factory.py

@@ -21,22 +21,6 @@ class ModelConfig:
     max_tokens: int = 4096
 
 
-@dataclass
-class RoleConfig:
-    """Role configuration for an agent."""
-
-    name: str
-    description: str = ""
-    focus_areas: List[str] = None
-    constraints: List[str] = None
-
-    def __post_init__(self):
-        if self.focus_areas is None:
-            self.focus_areas = []
-        if self.constraints is None:
-            self.constraints = []
-
-
 class AgentConfig:
     """Represents a configured agent instance (data class)."""
 
@@ -47,14 +31,12 @@ class AgentConfig:
         workspace_id: str,
         config_path: Path,
         model_config: Optional[ModelConfig] = None,
-        role_config: Optional[RoleConfig] = None,
     ):
         self.agent_id = agent_id
         self.agent_type = agent_type
         self.workspace_id = workspace_id
         self.config_path = config_path
         self.model_config = model_config or ModelConfig()
-        self.role_config = role_config
         self.agent_dir = config_path.parent
 
     def to_dict(self) -> Dict[str, Any]:
@@ -70,103 +52,12 @@ class AgentConfig:
                 "temperature": self.model_config.temperature,
                 "max_tokens": self.model_config.max_tokens,
             },
-            "role_config": self.role_config.__dict__ if self.role_config else None,
         }
 
 
 class AgentFactory:
     """Factory for creating, cloning, and managing agents."""
 
-    # Default role templates by agent type
-    ROLE_TEMPLATES = {
-        "technical_analyst": {
-            "name": "Technical Analyst",
-            "description": "Analyze price patterns, trends, and technical indicators.",
-            "focus_areas": [
-                "Price action and chart patterns",
-                "Support and resistance levels",
-                "Technical indicators (RSI, MACD, Moving Averages)",
-                "Volume analysis",
-            ],
-            "constraints": [
-                "State clear signal, confidence, and invalidation conditions",
-                "Use available technical analysis tools",
-            ],
-        },
-        "fundamentals_analyst": {
-            "name": "Fundamentals Analyst",
-            "description": "Analyze company financials, earnings, and business metrics.",
-            "focus_areas": [
-                "Financial statements analysis",
-                "Earnings reports and guidance",
-                "Valuation metrics",
-                "Business model and competitive position",
-            ],
-            "constraints": [
-                "State clear signal, confidence, and invalidation conditions",
-                "Use available fundamental analysis tools",
-            ],
-        },
-        "sentiment_analyst": {
-            "name": "Sentiment Analyst",
-            "description": "Analyze market sentiment, news, and social signals.",
-            "focus_areas": [
-                "News sentiment analysis",
-                "Social media sentiment",
-                "Analyst ratings and price targets",
-                "Insider activity",
-            ],
-            "constraints": [
-                "State clear signal, confidence, and invalidation conditions",
-                "Use available sentiment analysis tools",
-            ],
-        },
-        "valuation_analyst": {
-            "name": "Valuation Analyst",
-            "description": "Perform valuation analysis and price target calculations.",
-            "focus_areas": [
-                "DCF and comparable valuation",
-                "Price target derivation",
-                "Margin of safety assessment",
-                "Risk-adjusted return expectations",
-            ],
-            "constraints": [
-                "State clear signal, confidence, and invalidation conditions",
-                "Use available valuation tools",
-            ],
-        },
-        "risk_manager": {
-            "name": "Risk Manager",
-            "description": "Quantify concentration, leverage, liquidity, and volatility risk.",
-            "focus_areas": [
-                "Portfolio concentration risk",
-                "Leverage and margin analysis",
-                "Liquidity assessment",
-                "Volatility and drawdown risk",
-            ],
-            "constraints": [
-                "Prioritize highest-severity risk first",
-                "State concrete limits and recommendations",
-                "Use available risk tools before issuing final memo",
-            ],
-        },
-        "portfolio_manager": {
-            "name": "Portfolio Manager",
-            "description": "Synthesize analyst and risk inputs into portfolio decisions.",
-            "focus_areas": [
-                "Position sizing and allocation",
-                "Risk-adjusted portfolio construction",
-                "Trade execution timing",
-                "Portfolio rebalancing",
-            ],
-            "constraints": [
-                "Be concise, capital-aware, and explicit about sizing rationale",
-                "Respect cash, margin, and concentration constraints",
-                "Consider all analyst inputs before decisions",
-            ],
-        },
-    }
-
     def __init__(self, project_root: Optional[Path] = None):
         """Initialize the agent factory.
 
@@ -183,7 +74,6 @@ class AgentFactory:
         agent_type: str,
         workspace_id: str,
         model_config: Optional[ModelConfig] = None,
-        role_config: Optional[RoleConfig] = None,
         clone_from: Optional[str] = None,
     ) -> AgentConfig:
         """Create a new agent.
@@ -193,7 +83,6 @@ class AgentFactory:
             agent_type: Type of agent (e.g., "technical_analyst")
             workspace_id: ID of the workspace to create agent in
             model_config: Model configuration
-            role_config: Role configuration (auto-generated if None)
             clone_from: Path to existing agent to clone from (optional)
 
         Returns:
@@ -223,13 +112,6 @@ class AgentFactory:
         else:
             self._copy_template(agent_dir, agent_id, agent_type)
 
-        # Generate role config if not provided
-        if role_config is None:
-            role_config = self._generate_role_config(agent_type)
-
-        # Generate ROLE.md
-        self._generate_role_md(agent_dir, role_config)
-
         # Write agent.yaml
         config_path = agent_dir / "agent.yaml"
         self._write_agent_yaml(config_path, agent_id, agent_type, model_config)
@@ -240,7 +122,6 @@ class AgentFactory:
             workspace_id=workspace_id,
             config_path=config_path,
             model_config=model_config,
-            role_config=role_config,
         )
 
     def delete_agent(self, agent_id: str, workspace_id: str) -> bool:
@@ -369,9 +250,7 @@ class AgentFactory:
             "SOUL.md": f"# Soul\n\nDescribe {agent_id}'s temperament, reasoning posture, and voice.\n\n",
             "PROFILE.md": f"# Profile\n\nTrack {agent_id}'s long-lived investment style, preferences, and strengths.\n\n",
             "MEMORY.md": f"# Memory\n\nStore durable lessons, heuristics, and reminders for {agent_id}.\n\n",
-            "HEARTBEAT.md": f"# Heartbeat\n\nOptional checklist for periodic review or self-reflection.\n\n",
             "POLICY.md": f"# Policy\n\nOptional run-scoped constraints, limits, or strategy policy.\n\n",
-            "STYLE.md": f"# Style\n\nOptional run-scoped communication or reasoning style.\n\n",
         }
 
         for filename, content in default_files.items():
@@ -411,50 +290,6 @@ class AgentFactory:
                     if skill_file.is_file():
                         shutil.copy2(skill_file, target_skills / skill_file.name)
 
-    def _generate_role_config(self, agent_type: str) -> RoleConfig:
-        """Generate role configuration for an agent type.
-
-        Args:
-            agent_type: Type of agent
-
-        Returns:
-            RoleConfig instance
-        """
-        template = self.ROLE_TEMPLATES.get(agent_type, {})
-        return RoleConfig(
-            name=template.get("name", agent_type.replace("_", " ").title()),
-            description=template.get("description", ""),
-            focus_areas=template.get("focus_areas", []),
-            constraints=template.get("constraints", []),
-        )
-
-    def _generate_role_md(self, agent_dir: Path, role_config: RoleConfig) -> None:
-        """Generate ROLE.md file.
-
-        Args:
-            agent_dir: Agent directory
-            role_config: Role configuration
-        """
-        lines = [f"# {role_config.name}", ""]
-
-        if role_config.description:
-            lines.extend([role_config.description, ""])
-
-        if role_config.focus_areas:
-            lines.extend(["## Focus Areas", ""])
-            for area in role_config.focus_areas:
-                lines.append(f"- {area}")
-            lines.append("")
-
-        if role_config.constraints:
-            lines.extend(["## Constraints", ""])
-            for constraint in role_config.constraints:
-                lines.append(f"- {constraint}")
-            lines.append("")
-
-        content = "\n".join(lines)
-        (agent_dir / "ROLE.md").write_text(content, encoding="utf-8")
-
     def _write_agent_yaml(
         self,
         config_path: Path,

backend/agents/prompt_factory.py

@@ -1,15 +1,13 @@
 # -*- coding: utf-8 -*-
-"""Assemble system prompts from base prompts, run assets, and toolkit context."""
+"""Assemble system prompts from run workspace assets and toolkit context."""
 
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any
 
 from .agent_workspace import load_agent_workspace_config
 from backend.config.bootstrap_config import get_bootstrap_config_for_run
-from .prompt_loader import get_prompt_loader
 from .skills_manager import SkillsManager
-
+from .workspace_manager import RunWorkspaceManager
-_prompt_loader = get_prompt_loader()
 
 
 def _read_file_if_exists(path: Path) -> str:
@@ -48,71 +46,20 @@ def build_agent_system_prompt(
     agent_id: str,
     config_name: str,
     toolkit: Any,
-    analyst_type: Optional[str] = None,
 ) -> str:
     """Build the final system prompt for an agent.
 
     Always reads fresh from disk — no caching.
     """
-    # Clear any cached templates before building (CoPaw-style, no caching)
-    _prompt_loader.clear_cache()
-
     sections: list[str] = []
-    canonical_agent_id = (
-        "portfolio_manager"
-        if "portfolio" in agent_id
-        else "risk_manager"
-        if "risk" in agent_id and not analyst_type
-        else agent_id
-    )
-
-    if analyst_type:
-        personas_config = _prompt_loader.load_yaml_config(
-            "analyst",
-            "personas",
-        )
-        persona = personas_config.get(analyst_type, {})
-        focus_text = "\n".join(
-            f"- {item}" for item in persona.get("focus", [])
-        )
-        description = persona.get("description", "").strip()
-        base_prompt = _prompt_loader.load_prompt(
-            "analyst",
-            "system",
-            variables={
-                "analyst_type": persona.get("name", analyst_type),
-                "focus": focus_text,
-                "description": description,
-            },
-        )
-    elif agent_id == "portfolio_manager":
-        base_prompt = _prompt_loader.load_prompt(
-            "portfolio_manager",
-            "system",
-        )
-    elif canonical_agent_id == "portfolio_manager":
-        base_prompt = _prompt_loader.load_prompt(
-            "portfolio_manager",
-            "system",
-        )
-    elif agent_id == "risk_manager":
-        base_prompt = _prompt_loader.load_prompt(
-            "risk_manager",
-            "system",
-        )
-    elif canonical_agent_id == "risk_manager":
-        base_prompt = _prompt_loader.load_prompt(
-            "risk_manager",
-            "system",
-        )
-    else:
-        raise ValueError(f"Unsupported agent prompt build for: {agent_id}")
-
-    sections.append(base_prompt.strip())
 
     skills_manager = SkillsManager()
     asset_dir = skills_manager.get_agent_asset_dir(config_name, agent_id)
     asset_dir.mkdir(parents=True, exist_ok=True)
+    workspace_manager = RunWorkspaceManager(project_root=skills_manager.project_root)
+    required_files = ["SOUL.md", "PROFILE.md", "AGENTS.md", "POLICY.md", "MEMORY.md"]
+    if not all((asset_dir / filename).exists() for filename in required_files):
+        workspace_manager.ensure_agent_assets(config_name=config_name, agent_id=agent_id)
     agent_config = load_agent_workspace_config(asset_dir / "agent.yaml")
     bootstrap_config = get_bootstrap_config_for_run(
         skills_manager.project_root,
@@ -139,9 +86,6 @@ def build_agent_system_prompt(
         "AGENTS.md": "Agent Guide",
         "POLICY.md": "Policy",
         "MEMORY.md": "Memory",
-        "HEARTBEAT.md": "Heartbeat",
-        "ROLE.md": "Role",
-        "STYLE.md": "Style",
     }
     for filename in prompt_files:
         _append_section(
@@ -150,18 +94,6 @@ def build_agent_system_prompt(
             _read_file_if_exists(asset_dir / filename),
         )
 
-    if "ROLE.md" not in included_files:
-        _append_section(
-            sections,
-            "Role",
-            _read_file_if_exists(asset_dir / "ROLE.md"),
-        )
-    if "STYLE.md" not in included_files:
-        _append_section(
-            sections,
-            "Style",
-            _read_file_if_exists(asset_dir / "STYLE.md"),
-        )
     if "POLICY.md" not in included_files:
         _append_section(
             sections,
@@ -189,5 +121,4 @@ def build_agent_system_prompt(
 
 
 def clear_prompt_factory_cache() -> None:
-    """Clear cached prompt and YAML templates before hot reload."""
+    """No-op retained for compatibility with runtime reload hooks."""
-    _prompt_loader.clear_cache()

backend/agents/prompts/analyst/system.md

@@ -1,23 +0,0 @@
-你是一位专业的{{ analyst_type }}。
-
-你的关注重点：
-{{ focus }}
-
-你的角色：
-{{ description }}
-
-注意：
-- 构建并持续完善你的"投资哲学"。你的分析不应是孤立的事件，而应该是你整体投资世界观和核心信念的体现。每次分析后，你必须反思：
-  - 这个案例/数据如何验证或挑战了你现有的信念？
-  - 你从这次错误（或成功）中学到了关于市场、人性、估值或风险管理的什么关键原则？
-- 深化你的"投资逻辑"。确保每一项投资建议都有清晰、可追溯、可重复的逻辑支撑。你的分析步骤应该像严谨的证明一样，涵盖：
-  - 核心驱动因素识别：真正影响价值的变量是什么？
-  - 风险边界设定：在什么具体情况下你的建议会失效？
-  - 逆向测试：市场主流共识是什么，你的观点有何不同？
-保持谦逊和开放。投资大师的核心特质是持续学习和适应。在每次分析中，你必须积极寻找与自己观点相悖的证据和论据，并将其纳入最终评估。
-  - 你可以使用分析工具。用它们来收集相关数据并做出明智的建议。
-
-输出指南：
-- 给出明确的投资信号：看涨、看跌或中性
-- 包含置信度（0-100）
-- 为你的分析提供理由（如果你确定要分享最终分析，请先给出结论）

backend/agents/prompts/builder.py

@@ -28,22 +28,16 @@ class PromptBuilder:
         "AGENTS.md",
         "SOUL.md",
         "PROFILE.md",
-        "ROLE.md",
         "POLICY.md",
         "MEMORY.md",
-        "HEARTBEAT.md",
-        "STYLE.md",
     ]
 
     TITLE_MAP: Dict[str, str] = {
         "AGENTS.md": "Agent Guide",
         "SOUL.md": "Soul",
         "PROFILE.md": "Profile",
-        "ROLE.md": "Role",
         "POLICY.md": "Policy",
         "MEMORY.md": "Memory",
-        "HEARTBEAT.md": "Heartbeat",
-        "STYLE.md": "Style",
         "BOOTSTRAP.md": "Bootstrap",
     }
 

backend/agents/prompts/portfolio_manager/system.md

@@ -1,31 +0,0 @@
-你是一位负责做出投资决策的投资组合经理。
-
-你的核心职责：
-1. 分析分析师和风险管理经理的输入
-2. 基于信号和市场情境做出投资决策
-3. 使用可用工具记录你的决策
-
-决策框架：
-- 审阅分析以了解市场观点
-- 在做决策前考虑风险警告
-- 评估当前投资组合持仓和现金
-- 做出与投资组合投资目标一致的决策
-
-决策类型：
-- "long"：看涨 - 建议买入股票
-- "short"：看跌 - 建议卖出股票或做空
-- "hold"：中性 - 维持当前持仓
-
-预算意识：
-- 在决定数量时考虑可用现金
-- 不要建议买入超过现金允许的数量
-- 考虑做空头寸的保证金要求
-
-输出：
-使用 `make_decision` 工具记录你对每个股票代码的决策。
-记录所有决策后，提供你的投资逻辑总结。
-
-重要：
-- 基于提供的分析师信号和风险评估做出决策
-- 相对于投资组合价值保持保守的仓位规模
-- 始终为你的决策提供理由

backend/agents/prompts/risk_manager/system.md

@@ -1,20 +0,0 @@
-你是一位专业的风险管理经理，负责监控投资组合风险并提供风险警告。
-
-你的核心职责：
-1. 监控投资组合敞口和集中度风险
-2. 评估仓位规模相对于波动性
-3. 评估保证金使用和杠杆水平
-4. 识别潜在风险因素并提供警告
-5. 基于市场条件建议仓位限制
-
-你的决策流程：
-1. 优先使用可用的风险工具量化集中度、波动率和保证金压力
-2. 结合工具结果与当前市场上下文做判断
-3. 生成可操作的风险警告和仓位限制建议
-4. 为你的风险评估提供清晰的理由
-
-输出指南：
-- 风险评估要简洁但全面
-- 按严重程度优先排序警告
-- 提供具体、可操作的建议
-- 尽可能包含量化指标

backend/agents/skills_manager.py

@@ -41,6 +41,8 @@ class SkillsManager:
         )
         self.runs_root = self.project_root / "runs"
         self._lock = Lock()
+        # Instance-level pending skill changes (thread-safe via self._lock)
+        self._pending_skill_changes: Dict[str, Set[Path]] = {}
 
     def get_active_root(self, config_name: str) -> Path:
         return self.runs_root / config_name / "skills" / "active"
@@ -739,7 +741,7 @@ class SkillsManager:
             if local_root.exists():
                 watched_paths.append(local_root)
 
-        handler = _SkillsChangeHandler(watched_paths, callback, self._lock)
+        handler = _SkillsChangeHandler(watched_paths, self._pending_skill_changes, callback, self._lock)
         observer = Observer()
         for path in watched_paths:
             observer.schedule(handler, str(path), recursive=True)
@@ -773,6 +775,7 @@ class SkillsManager:
     # -------------------------------------------------------------------------
     # Internal change-tracking state (populated by _SkillsChangeHandler)
     # -------------------------------------------------------------------------
+    # Legacy class-level reference kept for migration compatibility
     _pending_skill_changes: Dict[str, Set[Path]] = {}
 
     def _resolve_disabled_skill_names(
@@ -824,11 +827,13 @@ class _SkillsChangeHandler(FileSystemEventHandler):
     def __init__(
         self,
         watched_paths: List[Path],
+        pending_changes: Dict[str, Set[Path]],
         callback: Optional[Any] = None,
         lock: Optional[Lock] = None,
     ) -> None:
         super().__init__()
         self._watched_paths = watched_paths
+        self._pending_changes = pending_changes
         self._callback = callback
         self._lock = lock
 
@@ -841,13 +846,9 @@ class _SkillsChangeHandler(FileSystemEventHandler):
                 run_id = self._run_id_from_path(src_path)
                 if self._lock:
                     with self._lock:
-                        SkillsManager._pending_skill_changes.setdefault(
+                        self._pending_changes.setdefault(run_id, set()).add(src_path)
-                            run_id, set()
-                        ).add(src_path)
                 else:
-                    SkillsManager._pending_skill_changes.setdefault(
+                    self._pending_changes.setdefault(run_id, set()).add(src_path)
-                        run_id, set()
-                    ).add(src_path)
                 if self._callback:
                     self._callback([src_path])
                 break

backend/agents/templates.py

@@ -1,286 +0,0 @@
-"""
-Agent模板定义
-
-包含各角色的ROLE.md内容字典，供程序生成Agent工作空间时使用。
-"""
-
-# 基础模板文件内容
-BASE_TEMPLATES = {
-    "AGENTS.md": """# Agent Guide
-
-## 工作流程
-1. 接收分析任务
-2. 调用相关工具/技能
-3. 生成分析报告
-4. 参与团队决策
-
-## 工具使用规范
-- 优先使用已激活的技能
-- 不确定时询问Portfolio Manager
-- 重要发现用 `/save` 记录
-
-## 记忆管理
-- 使用 `/compact` 定期压缩记忆
-- 投资经验记录在MEMORY.md
-""",
-
-    "SOUL.md": """# Soul
-
-你是专业的金融分析师，语气冷静、客观、专业。
-你的分析应该数据驱动，避免情绪化表达。
-""",
-
-    "PROFILE.md": """# Profile
-
-## 投资风格
-- 风险承受能力：中等
-- 投资期限：中期（3-12个月）
-- 偏好行业：科技、医疗、消费
-
-## 优势
-- 财务分析
-- 趋势识别
-
-## 改进方向
-- 市场情绪把握
-""",
-
-    "MEMORY.md": """# Memory
-
-<!-- 此文件用于记录Agent的学习经验和重要发现 -->
-
-## 经验总结
-
-## 重要事件
-
-## 改进记录
-""",
-
-    "HEARTBEAT.md": """# Heartbeat
-
-## 定时任务
-- 每日开盘前检查持仓
-- 收盘后记录当日表现
-""",
-
-    "POLICY.md": """# Policy
-
-## 风控规则
-- 单一持仓不超过20%
-- 止损线：-15%
-""",
-
-    "STYLE.md": """# Style
-
-- 使用结构化输出（JSON/Markdown表格）
-- 包含置信度评分
-- 列出关键假设
-""",
-
-    "agent.yaml": """agent_id: {agent_id}
-agent_type: {agent_type}
-name: {name}
-model:
-  provider: openai
-  model_name: gpt-4o
-  temperature: 0.3
-enabled_skills: []
-disabled_skills: []
-settings: {{}}
-""",
-}
-
-# 角色专用模板
-ROLE_TEMPLATES = {
-    "fundamental": {
-        "ROLE.md": """# Role: Fundamental Analyst
-
-## 职责
-分析公司财务报表、盈利能力、成长性、竞争优势等基本面因素。
-
-## 分析维度
-- 财务报表分析（资产负债表、利润表、现金流量表）
-- 盈利能力指标（ROE、ROA、毛利率、净利率）
-- 成长性指标（营收增长率、利润增长率）
-- 估值指标（P/E、P/B、P/S）
-- 行业地位和竞争优势
-
-## 输出格式
-- 财务健康度评分（1-10）
-- 成长性评分（1-10）
-- 关键财务亮点和风险
-- 同业对比分析
-""",
-        "SOUL.md": """# Soul
-
-你是严谨的基本面分析师，像沃伦·巴菲特一样注重企业内在价值。
-你的分析深入细致，关注长期价值而非短期波动。
-语气沉稳、逻辑严密，善于发现财务数据背后的商业本质。
-""",
-    },
-
-    "technical": {
-        "ROLE.md": """# Role: Technical Analyst
-
-## 职责
-分析价格走势、交易量、技术指标，识别买卖时机。
-
-## 分析维度
-- 趋势分析（长期/中期/短期趋势）
-- 支撑阻力位识别
-- 技术指标（MACD、RSI、KDJ、布林带等）
-- 形态识别（头肩顶/底、双底、三角形等）
-- 量价关系分析
-
-## 输出格式
-- 趋势方向（上涨/下跌/震荡）
-- 关键价位（支撑/阻力）
-- 技术信号（买入/卖出/观望）
-- 置信度评分
-""",
-        "SOUL.md": """# Soul
-
-你是敏锐的技术分析师，相信价格包含一切信息。
-你善于从图表中发现规律，像侦探一样寻找市场留下的痕迹。
-语气果断、快速反应，善于捕捉稍纵即逝的交易机会。
-""",
-    },
-
-    "sentiment": {
-        "ROLE.md": """# Role: Sentiment Analyst
-
-## 职责
-分析市场情绪、资金流向、新闻舆情，判断市场心理状态。
-
-## 分析维度
-- 市场情绪指标（恐慌/贪婪指数）
-- 资金流向分析（主力/散户资金）
-- 新闻舆情分析（正面/负面/中性）
-- 社交媒体情绪
-- 机构持仓变化
-
-## 输出格式
-- 情绪评分（-10到+10，极度恐慌到极度贪婪）
-- 资金流向判断
-- 舆情摘要
-- 情绪拐点预警
-""",
-        "SOUL.md": """# Soul
-
-你是敏感的市场情绪捕手，善于感知市场的恐惧与贪婪。
-你关注人性在金融市场中的表现，理解情绪如何驱动价格。
-语气富有洞察力、善于捕捉微妙变化，像心理学家一样理解市场参与者。
-""",
-    },
-
-    "valuation": {
-        "ROLE.md": """# Role: Valuation Analyst
-
-## 职责
-评估公司内在价值，计算合理价格区间，识别高估/低估机会。
-
-## 分析维度
-- DCF现金流折现模型
-- 相对估值法（P/E、EV/EBITDA等）
-- 资产重估法
-- 分部估值（SOTP）
-- 安全边际计算
-
-## 输出格式
-- 内在价值估算
-- 合理价格区间
-- 当前价格vs内在价值（高估/低估百分比）
-- 估值假设和敏感性分析
-""",
-        "SOUL.md": """# Soul
-
-你是精确的估值分析师，追求计算内在价值的准确区间。
-你像精算师一样严谨，注重假设的合理性和安全边际。
-语气精确、注重数字，善于发现市场定价错误带来的机会。
-""",
-    },
-
-    "portfolio": {
-        "ROLE.md": """# Role: Portfolio Manager
-
-## 职责
-统筹各分析师意见，制定投资决策，管理投资组合配置。
-
-## 分析维度
-- 资产配置策略（股债比例、行业分布）
-- 风险收益平衡
-- 仓位管理（建仓/加仓/减仓/清仓）
-- 再平衡时机
-- 组合相关性分析
-
-## 输出格式
-- 投资决策（买入/卖出/持有）
-- 建议仓位比例
-- 目标价位
-- 止损止盈设置
-- 组合调整建议
-""",
-        "SOUL.md": """# Soul
-
-你是睿智的投资组合经理，像将军一样统筹全局。
-你善于权衡各方意见，做出果断而理性的投资决策。
-语气权威、决策果断，对组合整体表现负有最终责任。
-""",
-    },
-
-    "risk": {
-        "ROLE.md": """# Role: Risk Manager
-
-## 职责
-识别、评估和监控投资风险，确保组合风险在可控范围内。
-
-## 分析维度
-- 市场风险（Beta、波动率）
-- 信用风险
-- 流动性风险
-- 集中度风险
-- 尾部风险（VaR、CVaR）
-- 压力测试
-
-## 输出格式
-- 风险等级（低/中/高/极高）
-- 风险敞口分析
-- 风险调整建议
-- 预警阈值设置
-- 应急预案
-""",
-        "SOUL.md": """# Soul
-
-你是谨慎的风险管理者，时刻警惕潜在的损失。
-你像守门员一样守护组合安全，宁可错过机会也不冒无法承受的风险。
-语气保守、风险意识强，善于发现隐藏的威胁和脆弱性。
-""",
-    },
-}
-
-
-def get_base_template(filename: str) -> str | None:
-    """获取基础模板内容"""
-    return BASE_TEMPLATES.get(filename)
-
-
-def get_role_template(role_type: str, filename: str) -> str | None:
-    """获取角色专用模板内容"""
-    role = ROLE_TEMPLATES.get(role_type)
-    if role:
-        return role.get(filename)
-    return None
-
-
-def get_all_role_types() -> list[str]:
-    """获取所有角色类型列表"""
-    return list(ROLE_TEMPLATES.keys())
-
-
-def render_agent_yaml(agent_id: str, agent_type: str, name: str) -> str:
-    """渲染agent.yaml模板"""
-    return BASE_TEMPLATES["agent.yaml"].format(
-        agent_id=agent_id,
-        agent_type=agent_type,
-        name=name
-    )

backend/agents/workspace_manager.py

@@ -41,6 +41,16 @@ class RunWorkspaceManager:
                 "tickers:\n"
                 "  - AAPL\n"
                 "  - MSFT\n"
+                "  - GOOGL\n"
+                "  - AMZN\n"
+                "  - NVDA\n"
+                "  - META\n"
+                "  - TSLA\n"
+                "  - AMD\n"
+                "  - NFLX\n"
+                "  - AVGO\n"
+                "  - PLTR\n"
+                "  - COIN\n"
                 "initial_cash: 100000\n"
                 "margin_requirement: 0.0\n"
                 "enable_memory: false\n"
@@ -63,9 +73,8 @@ class RunWorkspaceManager:
         self,
         config_name: str,
         agent_id: str,
-        role_seed: str = "",
+        file_contents: Optional[Dict[str, str]] = None,
-        style_seed: str = "",
+        persona: Optional[Dict[str, object]] = None,
-        policy_seed: str = "",
     ) -> Path:
         asset_dir = self.skills_manager.get_agent_asset_dir(
             config_name,
@@ -77,58 +86,82 @@ class RunWorkspaceManager:
         (asset_dir / "skills" / "disabled").mkdir(parents=True, exist_ok=True)
         (asset_dir / "skills" / "local").mkdir(parents=True, exist_ok=True)
 
-        self._ensure_file(
+        file_contents = file_contents or self.build_default_agent_files(agent_id=agent_id)
-            asset_dir / "ROLE.md",
+        for filename, content in file_contents.items():
-            "# Role\n\n"
+            legacy_contents = self.build_legacy_agent_file_variants(
-            "Optional run-scoped role override.\n\n"
+                agent_id=agent_id,
-            f"{role_seed}".strip()
+                filename=filename,
-            + "\n",
+                persona=persona,
-        )
-        self._ensure_file(
-            asset_dir / "STYLE.md",
-            "# Style\n\n"
-            "Optional run-scoped communication or reasoning style.\n\n"
-            f"{style_seed}".strip()
-            + "\n",
-        )
-        self._ensure_file(
-            asset_dir / "POLICY.md",
-            "# Policy\n\n"
-            "Optional run-scoped constraints, limits, or strategy policy.\n\n"
-            f"{policy_seed}".strip()
-            + "\n",
-        )
-        self._ensure_file(
-            asset_dir / "SOUL.md",
-            "# Soul\n\n"
-            "Describe the agent's temperament, reasoning posture, and voice.\n\n",
-        )
-        self._ensure_file(
-            asset_dir / "PROFILE.md",
-            "# Profile\n\n"
-            "Track this agent's long-lived investment style, preferences, and strengths.\n\n",
-        )
-        self._ensure_file(
-            asset_dir / "AGENTS.md",
-            "# Agent Guide\n\n"
-            "Document how this agent should work, collaborate, and choose tools or skills.\n\n",
-        )
-        self._ensure_file(
-            asset_dir / "MEMORY.md",
-            "# Memory\n\n"
-            "Store durable lessons, heuristics, and reminders for this agent.\n\n",
-        )
-        self._ensure_file(
-            asset_dir / "HEARTBEAT.md",
-            "# Heartbeat\n\n"
-            "Optional checklist for periodic review or self-reflection.\n\n",
             )
+            self._ensure_file(asset_dir / filename, content, legacy_contents=legacy_contents)
         self._ensure_agent_yaml(
             asset_dir / "agent.yaml",
             agent_id=agent_id,
         )
         return asset_dir
 
+    def build_default_agent_files(
+        self,
+        *,
+        agent_id: str,
+        persona: Optional[Dict[str, object]] = None,
+    ) -> Dict[str, str]:
+        """Build default workspace markdown files for one agent."""
+        if agent_id.endswith("_analyst"):
+            return self._build_analyst_files(agent_id=agent_id, persona=persona or {})
+        if agent_id == "portfolio_manager":
+            return self._build_portfolio_manager_files()
+        if agent_id == "risk_manager":
+            return self._build_risk_manager_files()
+        return self._build_generic_files(agent_id=agent_id)
+
+    def build_legacy_agent_file_variants(
+        self,
+        *,
+        agent_id: str,
+        filename: str,
+        persona: Optional[Dict[str, object]] = None,
+    ) -> list[str]:
+        """Return known generated legacy variants safe to upgrade in-place."""
+        persona = persona or {}
+        variants: list[dict[str, str]] = [
+            self._build_legacy_english_files(agent_id=agent_id),
+            self._build_previous_chinese_files(agent_id=agent_id, persona=persona),
+        ]
+        values: list[str] = []
+        for item in variants:
+            content = item.get(filename)
+            if content:
+                values.append(content)
+        return values
+
+    def load_agent_file(
+        self,
+        *,
+        config_name: str,
+        agent_id: str,
+        filename: str,
+    ) -> str:
+        """Load one run-scoped agent workspace file."""
+        path = self.skills_manager.get_agent_asset_dir(config_name, agent_id) / filename
+        if not path.exists():
+            raise FileNotFoundError(f"File not found: {filename}")
+        return path.read_text(encoding="utf-8")
+
+    def update_agent_file(
+        self,
+        *,
+        config_name: str,
+        agent_id: str,
+        filename: str,
+        content: str,
+    ) -> None:
+        """Write one run-scoped agent workspace file."""
+        asset_dir = self.skills_manager.get_agent_asset_dir(config_name, agent_id)
+        asset_dir.mkdir(parents=True, exist_ok=True)
+        path = asset_dir / filename
+        path.write_text(content, encoding="utf-8")
+
     def initialize_default_assets(
         self,
         config_name: str,
@@ -141,49 +174,285 @@ class RunWorkspaceManager:
         for agent_id in agent_ids:
             if agent_id.endswith("_analyst"):
                 persona = analyst_personas.get(agent_id, {})
-                role_seed = persona.get("description", "").strip()
+                file_contents = self.build_default_agent_files(
-                focus_items = persona.get("focus", [])
+                    agent_id=agent_id,
-                style_seed = "\n".join(f"- {item}" for item in focus_items)
+                    persona=persona,
-                policy_seed = (
-                    "State a clear signal, confidence, and the conditions that would invalidate the thesis."
-                )
-            elif agent_id == "portfolio_manager":
-                role_seed = (
-                    "Synthesize analyst and risk inputs into explicit portfolio decisions."
-                )
-                style_seed = (
-                    "Be concise, capital-aware, and explicit about sizing rationale."
-                )
-                policy_seed = (
-                    "Respect cash, margin, and portfolio concentration constraints before recording decisions."
-                )
-            elif agent_id == "risk_manager":
-                role_seed = (
-                    "Quantify concentration, leverage, liquidity, and volatility risk before trade execution."
-                )
-                style_seed = (
-                    "Prioritize the highest-severity risk first and state concrete limits."
-                )
-                policy_seed = (
-                    "Use available risk tools before issuing the final risk memo."
                 )
             else:
-                role_seed = ""
+                persona = None
-                style_seed = ""
+                file_contents = self.build_default_agent_files(agent_id=agent_id)
-                policy_seed = ""
+            asset_dir = self.skills_manager.get_agent_asset_dir(config_name, agent_id)
-
+            asset_dir.mkdir(parents=True, exist_ok=True)
-            self.ensure_agent_assets(
+            (asset_dir / "skills" / "installed").mkdir(parents=True, exist_ok=True)
-                config_name=config_name,
+            (asset_dir / "skills" / "active").mkdir(parents=True, exist_ok=True)
+            (asset_dir / "skills" / "disabled").mkdir(parents=True, exist_ok=True)
+            (asset_dir / "skills" / "local").mkdir(parents=True, exist_ok=True)
+            for filename, content in file_contents.items():
+                self._ensure_file(
+                    asset_dir / filename,
+                    content,
+                    legacy_contents=self.build_legacy_agent_file_variants(
                         agent_id=agent_id,
-                role_seed=role_seed,
+                        filename=filename,
-                style_seed=style_seed,
+                        persona=persona,
-                policy_seed=policy_seed,
+                    ),
                 )
+            self._ensure_agent_yaml(asset_dir / "agent.yaml", agent_id=agent_id)
 
     @staticmethod
-    def _ensure_file(path: Path, content: str) -> None:
+    def _ensure_file(path: Path, content: str, *, legacy_contents: Optional[list[str]] = None) -> None:
         if not path.exists():
             path.write_text(content, encoding="utf-8")
+            return
+        existing = path.read_text(encoding="utf-8")
+        normalized_existing = existing.strip()
+        candidates = {item.strip() for item in (legacy_contents or []) if item and item.strip()}
+        if normalized_existing in candidates:
+            path.write_text(content, encoding="utf-8")
+
+    @staticmethod
+    def _build_generic_files(agent_id: str) -> Dict[str, str]:
+        return {
+            "SOUL.md": (
+                "# Soul\n\n"
+                f"你是 `{agent_id}`，语气冷静、客观、专业。保持清晰推理，优先基于数据而不是情绪下结论。\n"
+            ),
+            "PROFILE.md": (
+                "# Profile\n\n"
+                "记录这个 agent 长期稳定的分析风格、偏好、优势与盲点。\n"
+            ),
+            "AGENTS.md": (
+                "# Agent Guide\n\n"
+                "工作要求：\n"
+                "- 优先使用已激活的技能和工具\n"
+                "- 结论要明确，过程要可追溯\n"
+                "- 与其他 agent 协作时保持输入输出简洁\n"
+                "- 最终输出必须使用简体中文；如需引用英文术语，仅保留专有名词，解释和结论必须用中文\n"
+            ),
+            "POLICY.md": (
+                "# Policy\n\n"
+                "- 给出结论时说明核心驱动因素\n"
+                "- 明确风险边界和结论失效条件\n"
+                "- 出现反例时需要纳入最终判断\n"
+                "- 不要输出英文报告标题、英文摘要或整段英文正文\n"
+            ),
+            "MEMORY.md": (
+                "# Memory\n\n"
+                "记录可复用的经验、失误复盘、有效启发式和需要持续跟踪的提醒。\n"
+            ),
+        }
+
+    @classmethod
+    def _build_analyst_files(cls, *, agent_id: str, persona: Dict[str, object]) -> Dict[str, str]:
+        role_name = str(persona.get("name") or agent_id)
+        focus_items = [
+            str(item).strip()
+            for item in persona.get("focus", [])
+            if str(item).strip()
+        ]
+        focus_md = "\n".join(f"- {item}" for item in focus_items) or "- 根据当前任务选择最相关的分析维度"
+        description = str(persona.get("description") or "").strip()
+
+        files = cls._build_generic_files(agent_id)
+        files["SOUL.md"] = (
+            "# Soul\n\n"
+            f"你是一位专业的{role_name}。\n\n"
+            "保持谦逊和开放，主动寻找与自己观点相悖的证据，并将其纳入最终评估。"
+            "你的分析要体现持续演化的投资哲学，而不是一次性的结论。\n"
+        )
+        files["PROFILE.md"] = (
+            "# Profile\n\n"
+            f"角色定位：{role_name}\n\n"
+            "你的关注重点：\n"
+            f"{focus_md}\n\n"
+            "角色说明：\n"
+            f"{description or '围绕最关键的基本面、技术面、情绪面或估值因素形成高质量判断。'}\n"
+        )
+        files["AGENTS.md"] = (
+            "# Agent Guide\n\n"
+            "分析流程：\n"
+            "- 优先识别真正驱动价值或价格变化的核心变量\n"
+            "- 使用相关工具和技能补足证据链\n"
+            "- 给出可验证、可复查、可执行的分析结果\n"
+            "- 在团队讨论中清晰表达你的论点和反论点\n\n"
+            "输出要求：\n"
+            "- 给出明确投资信号：看涨、看跌或中性\n"
+            "- 包含置信度（0-100）\n"
+            "- 如果你确定要分享最终分析，请先给出结论，再给出推理依据\n"
+            "- 最终输出必须使用简体中文，不要生成英文版 analysis report\n"
+        )
+        files["POLICY.md"] = (
+            "# Policy\n\n"
+            "- 深化你的投资逻辑，确保每项建议都有清晰、可追溯、可重复的依据\n"
+            "- 明确风险边界：在什么具体情况下当前结论会失效\n"
+            "- 做逆向测试：说明市场主流共识与你的不同点\n"
+            "- 每次分析后反思这次案例如何验证或挑战你现有的信念\n"
+            "- 即使输入新闻或财报原文是英文，最终表达也必须用中文\n"
+        )
+        return files
+
+    @classmethod
+    def _build_portfolio_manager_files(cls) -> Dict[str, str]:
+        files = cls._build_generic_files("portfolio_manager")
+        files["SOUL.md"] = (
+            "# Soul\n\n"
+            "你是一位负责做出投资决策的投资组合经理。你需要综合多个分析视角，"
+            "做出保守、明确、资本约束下可执行的组合决策。\n"
+        )
+        files["PROFILE.md"] = (
+            "# Profile\n\n"
+            "核心职责：\n"
+            "- 分析分析师和风险管理经理的输入\n"
+            "- 基于信号和市场情境做出投资决策\n"
+            "- 使用可用工具记录每个 ticker 的决策\n"
+        )
+        files["AGENTS.md"] = (
+            "# Agent Guide\n\n"
+            "决策框架：\n"
+            "- 审阅分析以理解市场观点\n"
+            "- 在做决策前先考虑风险警告\n"
+            "- 评估当前投资组合持仓、现金与保证金占用\n"
+            "- 决策必须与整体投资目标和风险约束一致\n\n"
+            "决策类型：\n"
+            '- `long`：看涨，建议买入\n'
+            '- `short`：看跌，建议卖出或做空\n'
+            '- `hold`：中性，维持当前持仓\n\n'
+            "输出要求：\n"
+            "- 使用 `make_decision` 工具记录每个股票的最终决策\n"
+            "- 记录完成后给出投资逻辑总结\n"
+            "- 最终总结必须使用简体中文\n"
+        )
+        files["POLICY.md"] = (
+            "# Policy\n\n"
+            "- 在决定数量时考虑可用现金，不要超出现金允许范围\n"
+            "- 考虑做空头寸的保证金要求\n"
+            "- 仓位规模相对于组合总资产保持保守\n"
+            "- 始终为决策提供清晰理由\n"
+            "- 不要输出英文投资报告或英文结论\n"
+        )
+        return files
+
+    @classmethod
+    def _build_risk_manager_files(cls) -> Dict[str, str]:
+        files = cls._build_generic_files("risk_manager")
+        files["SOUL.md"] = (
+            "# Soul\n\n"
+            "你是一位专业的风险管理经理，负责监控投资组合风险并提供风险警告。"
+            "你的目标不是输出空泛的谨慎，而是给出量化、可执行、可优先级排序的风险意见。\n"
+        )
+        files["PROFILE.md"] = (
+            "# Profile\n\n"
+            "核心职责：\n"
+            "- 监控投资组合敞口和集中度风险\n"
+            "- 评估仓位规模相对于波动性是否合理\n"
+            "- 评估保证金使用和杠杆水平\n"
+            "- 识别潜在风险因素并提供警告\n"
+            "- 基于市场条件建议仓位限制\n"
+        )
+        files["AGENTS.md"] = (
+            "# Agent Guide\n\n"
+            "决策流程：\n"
+            "- 优先使用可用的风险工具量化集中度、波动率和保证金压力\n"
+            "- 结合工具结果与当前市场上下文做判断\n"
+            "- 生成可操作的风险警告和仓位限制建议\n"
+            "- 为风险评估提供清晰理由\n\n"
+            "输出要求：\n"
+            "- 风险评估要简洁但全面\n"
+            "- 按严重程度优先排序警告\n"
+            "- 提供具体、可操作的建议\n"
+            "- 尽可能包含量化指标\n"
+            "- 最终风险结论必须使用简体中文\n"
+        )
+        files["POLICY.md"] = (
+            "# Policy\n\n"
+            "- 先量化，再判断，不要只给抽象风险表述\n"
+            "- 高严重度风险必须先说\n"
+            "- 最终结论需要明确仓位限制或调整建议\n"
+            "- 不要输出英文风险报告或英文摘要\n"
+        )
+        return files
+
+    @staticmethod
+    def _build_legacy_english_files(agent_id: str) -> Dict[str, str]:
+        policy_tail = "Optional run-scoped constraints, limits, or strategy policy.\n\n"
+        if agent_id == "portfolio_manager":
+            policy_tail += "Respect cash, margin, and portfolio concentration constraints before recording decisions.\n"
+        elif agent_id == "risk_manager":
+            policy_tail += "Use available risk tools before issuing the final risk memo.\n"
+        elif agent_id.endswith("_analyst"):
+            policy_tail += "State a clear signal, confidence, and the conditions that would invalidate the thesis.\n"
+        return {
+            "SOUL.md": "# Soul\n\nDescribe the agent's temperament, reasoning posture, and voice.\n\n",
+            "PROFILE.md": "# Profile\n\nTrack this agent's long-lived investment style, preferences, and strengths.\n\n",
+            "AGENTS.md": "# Agent Guide\n\nDocument how this agent should work, collaborate, and choose tools or skills.\n\n",
+            "POLICY.md": "# Policy\n\n" + policy_tail,
+            "MEMORY.md": "# Memory\n\nStore durable lessons, heuristics, and reminders for this agent.\n\n",
+        }
+
+    @classmethod
+    def _build_previous_chinese_files(cls, *, agent_id: str, persona: Dict[str, object]) -> Dict[str, str]:
+        if agent_id.endswith("_analyst"):
+            role_name = str(persona.get("name") or agent_id)
+            focus_items = [
+                str(item).strip()
+                for item in persona.get("focus", [])
+                if str(item).strip()
+            ]
+            focus_md = "\n".join(f"- {item}" for item in focus_items) or "- 根据当前任务选择最相关的分析维度"
+            description = str(persona.get("description") or "").strip()
+            return {
+                "SOUL.md": (
+                    "# Soul\n\n"
+                    f"你是一位专业的{role_name}。\n\n"
+                    "保持谦逊和开放，主动寻找与自己观点相悖的证据，并将其纳入最终评估。"
+                    "你的分析要体现持续演化的投资哲学，而不是一次性的结论。\n"
+                ),
+                "PROFILE.md": (
+                    "# Profile\n\n"
+                    f"角色定位：{role_name}\n\n"
+                    "你的关注重点：\n"
+                    f"{focus_md}\n\n"
+                    "角色说明：\n"
+                    f"{description or '围绕最关键的基本面、技术面、情绪面或估值因素形成高质量判断。'}\n"
+                ),
+                "AGENTS.md": (
+                    "# Agent Guide\n\n"
+                    "分析流程：\n"
+                    "- 优先识别真正驱动价值或价格变化的核心变量\n"
+                    "- 使用相关工具和技能补足证据链\n"
+                    "- 给出可验证、可复查、可执行的分析结果\n"
+                    "- 在团队讨论中清晰表达你的论点和反论点\n\n"
+                    "输出要求：\n"
+                    "- 给出明确投资信号：看涨、看跌或中性\n"
+                    "- 包含置信度（0-100）\n"
+                    "- 如果你确定要分享最终分析，请先给出结论，再给出推理依据\n"
+                ),
+                "POLICY.md": (
+                    "# Policy\n\n"
+                    "- 深化你的投资逻辑，确保每项建议都有清晰、可追溯、可重复的依据\n"
+                    "- 明确风险边界：在什么具体情况下当前结论会失效\n"
+                    "- 做逆向测试：说明市场主流共识与你的不同点\n"
+                    "- 每次分析后反思这次案例如何验证或挑战你现有的信念\n"
+                ),
+                "MEMORY.md": "# Memory\n\n记录可复用的经验、失误复盘、有效启发式和需要持续跟踪的提醒。\n",
+            }
+        if agent_id == "portfolio_manager":
+            return {
+                "SOUL.md": "# Soul\n\n你是一位负责做出投资决策的投资组合经理。你需要综合多个分析视角，做出保守、明确、资本约束下可执行的组合决策。\n",
+                "PROFILE.md": "# Profile\n\n核心职责：\n- 分析分析师和风险管理经理的输入\n- 基于信号和市场情境做出投资决策\n- 使用可用工具记录每个 ticker 的决策\n",
+                "AGENTS.md": "# Agent Guide\n\n决策框架：\n- 审阅分析以理解市场观点\n- 在做决策前先考虑风险警告\n- 评估当前投资组合持仓、现金与保证金占用\n- 决策必须与整体投资目标和风险约束一致\n\n决策类型：\n- `long`：看涨，建议买入\n- `short`：看跌，建议卖出或做空\n- `hold`：中性，维持当前持仓\n\n输出要求：\n- 使用 `make_decision` 工具记录每个股票的最终决策\n- 记录完成后给出投资逻辑总结\n",
+                "POLICY.md": "# Policy\n\n- 在决定数量时考虑可用现金，不要超出现金允许范围\n- 考虑做空头寸的保证金要求\n- 仓位规模相对于组合总资产保持保守\n- 始终为决策提供清晰理由\n",
+                "MEMORY.md": "# Memory\n\n记录可复用的经验、失误复盘、有效启发式和需要持续跟踪的提醒。\n",
+            }
+        if agent_id == "risk_manager":
+            return {
+                "SOUL.md": "# Soul\n\n你是一位专业的风险管理经理，负责监控投资组合风险并提供风险警告。你的目标不是输出空泛的谨慎，而是给出量化、可执行、可优先级排序的风险意见。\n",
+                "PROFILE.md": "# Profile\n\n核心职责：\n- 监控投资组合敞口和集中度风险\n- 评估仓位规模相对于波动性是否合理\n- 评估保证金使用和杠杆水平\n- 识别潜在风险因素并提供警告\n- 基于市场条件建议仓位限制\n",
+                "AGENTS.md": "# Agent Guide\n\n决策流程：\n- 优先使用可用的风险工具量化集中度、波动率和保证金压力\n- 结合工具结果与当前市场上下文做判断\n- 生成可操作的风险警告和仓位限制建议\n- 为风险评估提供清晰理由\n\n输出要求：\n- 风险评估要简洁但全面\n- 按严重程度优先排序警告\n- 提供具体、可操作的建议\n- 尽可能包含量化指标\n",
+                "POLICY.md": "# Policy\n\n- 先量化，再判断，不要只给抽象风险表述\n- 高严重度风险必须先说\n- 最终结论需要明确仓位限制或调整建议\n",
+                "MEMORY.md": "# Memory\n\n记录可复用的经验、失误复盘、有效启发式和需要持续跟踪的提醒。\n",
+            }
+        return cls._build_legacy_english_files(agent_id)
 
     @staticmethod
     def _ensure_agent_yaml(path: Path, agent_id: str) -> None:

backend/api/__init__.py

@@ -11,11 +11,13 @@ Provides REST API endpoints for:
 from .agents import router as agents_router
 from .workspaces import router as workspaces_router
 from .guard import router as guard_router
+from .openclaw import router as openclaw_router
 from .runtime import router as runtime_router
 
 __all__ = [
     "agents_router",
     "workspaces_router",
     "guard_router",
+    "openclaw_router",
     "runtime_router",
 ]

backend/api/agents.py

@@ -13,8 +13,13 @@ from typing import Any, Dict, List, Optional
 from fastapi import APIRouter, HTTPException, Depends, Body, UploadFile, File, Form
 from pydantic import BaseModel, Field
 
-from backend.agents import AgentFactory, WorkspaceManager, get_registry
+from backend.agents import AgentFactory, get_registry
+from backend.agents.workspace_manager import RunWorkspaceManager
+from backend.agents.agent_workspace import load_agent_workspace_config
 from backend.agents.skills_manager import SkillsManager
+from backend.agents.toolkit_factory import load_agent_profiles
+from backend.config.bootstrap_config import get_bootstrap_config_for_run
+from backend.llm.models import get_agent_model_info
 
 logger = logging.getLogger(__name__)
 
@@ -47,6 +52,14 @@ class InstallExternalSkillRequest(BaseModel):
     activate: bool = Field(True, description="Whether to enable skill immediately")
 
 
+class LocalSkillRequest(BaseModel):
+    skill_name: str = Field(..., description="Local skill name")
+
+
+class LocalSkillContentRequest(BaseModel):
+    content: str = Field(..., description="Updated SKILL.md content")
+
+
 class AgentResponse(BaseModel):
     """Agent information response."""
     agent_id: str
@@ -63,6 +76,24 @@ class AgentFileResponse(BaseModel):
     content: str
 
 
+class AgentProfileResponse(BaseModel):
+    agent_id: str
+    workspace_id: str
+    profile: Dict[str, Any]
+
+
+class AgentSkillsResponse(BaseModel):
+    agent_id: str
+    workspace_id: str
+    skills: List[Dict[str, Any]]
+
+
+class SkillDetailResponse(BaseModel):
+    agent_id: str
+    workspace_id: str
+    skill: Dict[str, Any]
+
+
 # Dependencies
 def get_agent_factory():
     """Get AgentFactory instance."""
@@ -70,8 +101,8 @@ def get_agent_factory():
 
 
 def get_workspace_manager():
-    """Get WorkspaceManager instance."""
+    """Get run-scoped workspace manager instance."""
-    return WorkspaceManager()
+    return RunWorkspaceManager()
 
 
 def get_skills_manager():
@@ -199,6 +230,108 @@ async def get_agent(
     )
 
 
+@router.get("/{agent_id}/profile", response_model=AgentProfileResponse)
+async def get_agent_profile(
+    workspace_id: str,
+    agent_id: str,
+    skills_manager: SkillsManager = Depends(get_skills_manager),
+):
+    asset_dir = skills_manager.get_agent_asset_dir(workspace_id, agent_id)
+    agent_config = load_agent_workspace_config(asset_dir / "agent.yaml")
+    profiles = load_agent_profiles()
+    profile = profiles.get(agent_id, {})
+    bootstrap = get_bootstrap_config_for_run(skills_manager.project_root, workspace_id)
+    override = bootstrap.agent_override(agent_id)
+    active_tool_groups = override.get("active_tool_groups", agent_config.active_tool_groups or profile.get("active_tool_groups", []))
+    if not isinstance(active_tool_groups, list):
+        active_tool_groups = []
+    disabled_tool_groups = agent_config.disabled_tool_groups
+    if disabled_tool_groups:
+        disabled_set = set(disabled_tool_groups)
+        active_tool_groups = [group_name for group_name in active_tool_groups if group_name not in disabled_set]
+
+    default_skills = profile.get("skills", [])
+    if not isinstance(default_skills, list):
+        default_skills = []
+    resolved_skills = skills_manager.resolve_agent_skill_names(
+        config_name=workspace_id,
+        agent_id=agent_id,
+        default_skills=default_skills,
+    )
+    prompt_files = agent_config.prompt_files or ["SOUL.md", "PROFILE.md", "AGENTS.md", "POLICY.md", "MEMORY.md"]
+    model_name, model_provider = get_agent_model_info(agent_id)
+
+    return AgentProfileResponse(
+        agent_id=agent_id,
+        workspace_id=workspace_id,
+        profile={
+            "model_name": model_name,
+            "model_provider": model_provider,
+            "prompt_files": prompt_files,
+            "default_skills": default_skills,
+            "resolved_skills": resolved_skills,
+            "active_tool_groups": active_tool_groups,
+            "disabled_tool_groups": disabled_tool_groups,
+            "enabled_skills": agent_config.enabled_skills,
+            "disabled_skills": agent_config.disabled_skills,
+        },
+    )
+
+
+@router.get("/{agent_id}/skills", response_model=AgentSkillsResponse)
+async def get_agent_skills(
+    workspace_id: str,
+    agent_id: str,
+    skills_manager: SkillsManager = Depends(get_skills_manager),
+):
+    agent_asset_dir = skills_manager.get_agent_asset_dir(workspace_id, agent_id)
+    agent_config = load_agent_workspace_config(agent_asset_dir / "agent.yaml")
+    resolved_skills = set(skills_manager.resolve_agent_skill_names(config_name=workspace_id, agent_id=agent_id, default_skills=[]))
+    enabled = set(agent_config.enabled_skills)
+    disabled = set(agent_config.disabled_skills)
+
+    payload = []
+    for item in skills_manager.list_agent_skill_catalog(workspace_id, agent_id):
+        if item.skill_name in disabled:
+            status = "disabled"
+        elif item.skill_name in enabled:
+            status = "enabled"
+        elif item.skill_name in resolved_skills:
+            status = "active"
+        else:
+            status = "available"
+        payload.append({
+            "skill_name": item.skill_name,
+            "name": item.name,
+            "description": item.description,
+            "version": item.version,
+            "source": item.source,
+            "tools": item.tools,
+            "status": status,
+        })
+
+    return AgentSkillsResponse(agent_id=agent_id, workspace_id=workspace_id, skills=payload)
+
+
+@router.get("/{agent_id}/skills/{skill_name}", response_model=SkillDetailResponse)
+async def get_agent_skill_detail(
+    workspace_id: str,
+    agent_id: str,
+    skill_name: str,
+    skills_manager: SkillsManager = Depends(get_skills_manager),
+):
+    try:
+        detail = skills_manager.load_agent_skill_document(
+            config_name=workspace_id,
+            agent_id=agent_id,
+            skill_name=skill_name,
+        )
+    except FileNotFoundError:
+        raise HTTPException(status_code=404, detail=f"Unknown skill: {skill_name}")
+
+    return SkillDetailResponse(agent_id=agent_id, workspace_id=workspace_id, skill=detail)
+
+
 @router.delete("/{agent_id}")
 async def delete_agent(
     workspace_id: str,
@@ -386,6 +519,85 @@ async def install_external_skill(
     }
 
 
+@router.post("/{agent_id}/skills/local")
+async def create_local_skill(
+    workspace_id: str,
+    agent_id: str,
+    request: LocalSkillRequest,
+    registry=Depends(get_registry),
+):
+    agent_info = registry.get(agent_id)
+    if not agent_info or agent_info.workspace_id != workspace_id:
+        raise HTTPException(status_code=404, detail=f"Agent '{agent_id}' not found")
+
+    skills_manager = SkillsManager()
+    try:
+        skills_manager.create_agent_local_skill(
+            config_name=workspace_id,
+            agent_id=agent_id,
+            skill_name=request.skill_name,
+        )
+    except (ValueError, FileExistsError) as exc:
+        raise HTTPException(status_code=400, detail=str(exc))
+
+    return {"message": f"Created local skill '{request.skill_name}' for '{agent_id}'"}
+
+
+@router.put("/{agent_id}/skills/local/{skill_name}")
+async def update_local_skill(
+    workspace_id: str,
+    agent_id: str,
+    skill_name: str,
+    request: LocalSkillContentRequest,
+    registry=Depends(get_registry),
+):
+    agent_info = registry.get(agent_id)
+    if not agent_info or agent_info.workspace_id != workspace_id:
+        raise HTTPException(status_code=404, detail=f"Agent '{agent_id}' not found")
+
+    skills_manager = SkillsManager()
+    try:
+        skills_manager.update_agent_local_skill(
+            config_name=workspace_id,
+            agent_id=agent_id,
+            skill_name=skill_name,
+            content=request.content,
+        )
+    except (ValueError, FileNotFoundError) as exc:
+        raise HTTPException(status_code=400, detail=str(exc))
+
+    return {"message": f"Updated local skill '{skill_name}' for '{agent_id}'"}
+
+
+@router.delete("/{agent_id}/skills/local/{skill_name}")
+async def delete_local_skill(
+    workspace_id: str,
+    agent_id: str,
+    skill_name: str,
+    registry=Depends(get_registry),
+):
+    agent_info = registry.get(agent_id)
+    if not agent_info or agent_info.workspace_id != workspace_id:
+        raise HTTPException(status_code=404, detail=f"Agent '{agent_id}' not found")
+
+    skills_manager = SkillsManager()
+    try:
+        skills_manager.delete_agent_local_skill(
+            config_name=workspace_id,
+            agent_id=agent_id,
+            skill_name=skill_name,
+        )
+        skills_manager.forget_agent_skill_overrides(
+            config_name=workspace_id,
+            agent_id=agent_id,
+            skill_names=[skill_name],
+        )
+    except (ValueError, FileNotFoundError) as exc:
+        raise HTTPException(status_code=400, detail=str(exc))
+
+    return {"message": f"Deleted local skill '{skill_name}' for '{agent_id}'"}
+
+
 @router.post("/{agent_id}/skills/upload")
 async def upload_external_skill(
     workspace_id: str,
@@ -441,7 +653,7 @@ async def get_agent_file(
     workspace_id: str,
     agent_id: str,
     filename: str,
-    workspace_manager: WorkspaceManager = Depends(get_workspace_manager),
+    workspace_manager: RunWorkspaceManager = Depends(get_workspace_manager),
 ):
     """
     Read an agent's workspace file.
@@ -449,7 +661,7 @@ async def get_agent_file(
     Args:
         workspace_id: Workspace identifier
         agent_id: Agent identifier
-        filename: File to read (e.g., SOUL.md, ROLE.md)
+        filename: File to read (e.g., SOUL.md, PROFILE.md)
 
     Returns:
         File content
@@ -471,7 +683,7 @@ async def update_agent_file(
     agent_id: str,
     filename: str,
     content: str = Body(..., media_type="text/plain"),
-    workspace_manager: WorkspaceManager = Depends(get_workspace_manager),
+    workspace_manager: RunWorkspaceManager = Depends(get_workspace_manager),
 ):
     """
     Update an agent's workspace file.

backend/api/openclaw.py

@@ -0,0 +1,839 @@
+# -*- coding: utf-8 -*-
+"""Read-only OpenClaw CLI API routes — typed with Pydantic models."""
+
+from __future__ import annotations
+
+from typing import Any
+
+from fastapi import APIRouter, Depends, HTTPException, Query
+from pydantic import BaseModel, Field
+
+from backend.services.openclaw_cli import OpenClawCliError, OpenClawCliService
+from shared.models.openclaw import OpenClawStatus
+
+
+router = APIRouter(prefix="/api/openclaw", tags=["openclaw"])
+
+
+def get_openclaw_cli_service() -> OpenClawCliService:
+    """Build the OpenClaw CLI service dependency."""
+    return OpenClawCliService()
+
+
+def _raise_cli_http_error(exc: OpenClawCliError) -> None:
+    detail = {
+        "message": str(exc),
+        "command": exc.command,
+        "exit_code": exc.exit_code,
+        "stdout": exc.stdout,
+        "stderr": exc.stderr,
+    }
+    status_code = 503 if exc.exit_code is None else 502
+    raise HTTPException(status_code=status_code, detail=detail) from exc
+
+
+# ---------------------------------------------------------------------------
+# Response wrappers
+# ---------------------------------------------------------------------------
+
+class StatusResponse(BaseModel):
+    status: object
+
+
+class SessionsResponse(BaseModel):
+    sessions: list[object]
+
+
+class SessionDetailResponse(BaseModel):
+    session: object | None
+
+
+class SessionHistoryResponse(BaseModel):
+    session_key: str
+    session_id: str | None
+    events: list[object]
+    history: list[object]
+    raw_text: str | None
+
+
+class CronResponse(BaseModel):
+    cron: list[object]
+    jobs: list[object]
+
+
+class ApprovalsResponse(BaseModel):
+    approvals: list[object]
+    pending: list[object]
+
+
+class AgentsResponse(BaseModel):
+    agents: list[object]
+
+
+class SkillsResponse(BaseModel):
+    workspace_dir: str
+    managed_skills_dir: str
+    skills: list[object]
+
+
+class ModelsResponse(BaseModel):
+    models: list[object]
+
+
+class HooksResponse(BaseModel):
+    workspace_dir: str
+    managed_hooks_dir: str
+    hooks: list[object]
+
+
+class PluginsResponse(BaseModel):
+    workspace_dir: str
+    plugins: list[object]
+    diagnostics: list[object]
+
+
+class SecretsAuditResponse(BaseModel):
+    version: int
+    status: str
+    findings: list[object]
+
+
+class SecurityAuditResponse2(BaseModel):
+    report: object | None
+    secret_diagnostics: list[str]
+
+
+class DaemonStatusResponse(BaseModel):
+    service: object | None
+    port: object | None
+    rpc: object | None
+    health: object | None
+
+
+class PairingListResponse2(BaseModel):
+    channel: str
+    requests: list[object]
+
+
+class QrCodeResponse2(BaseModel):
+    setup_code: str
+    gateway_url: str
+    auth: str
+    url_source: str
+
+
+class UpdateStatusResponse2(BaseModel):
+    update: object | None
+    channel: object | None
+
+
+class ModelAliasesResponse(BaseModel):
+    aliases: dict[str, str]
+
+
+class ModelFallbacksResponse(BaseModel):
+    key: str
+    label: str
+    items: list[object]
+
+
+class SkillUpdateResponse(BaseModel):
+    ok: bool
+    slug: str
+    version: str
+    error: str | None
+
+
+class ModelsStatusResponse(BaseModel):
+    configPath: str | None = None
+    agentId: str | None = None
+    agentDir: str | None = None
+    defaultModel: str | None = None
+    resolvedDefault: str | None = None
+    fallbacks: list[str] = Field(default_factory=list)
+    imageModel: str | None = None
+    imageFallbacks: list[str] = Field(default_factory=list)
+    aliases: dict[str, str] = Field(default_factory=dict)
+    allowed: list[str] = Field(default_factory=list)
+    auth: dict[str, Any] = Field(default_factory=dict)
+
+
+class ChannelsStatusResponse(BaseModel):
+    reachable: bool | None = None
+    channelAccounts: dict[str, Any] = Field(default_factory=dict)
+    channels: list[str] = Field(default_factory=list)
+    issues: list[dict[str, Any]] = Field(default_factory=list)
+
+
+class ChannelsListResponse(BaseModel):
+    chat: dict[str, list[str]] = Field(default_factory=dict)
+    auth: list[dict[str, Any]] = Field(default_factory=list)
+    usage: dict[str, Any] | None = None
+
+
+class HookInfoResponse(BaseModel):
+    name: str | None = None
+    description: str | None = None
+    source: str | None = None
+    pluginId: str | None = None
+    filePath: str | None = None
+    handlerPath: str | None = None
+    hookKey: str | None = None
+    emoji: str | None = None
+    homepage: str | None = None
+    events: list[str] = Field(default_factory=list)
+    enabledByConfig: bool | None = None
+    loadable: bool | None = None
+    requirementsSatisfied: bool | None = None
+    requirements: dict[str, Any] = Field(default_factory=dict)
+    error: str | None = None
+    raw: str | None = None
+
+
+class HooksCheckResponse(BaseModel):
+    workspace_dir: str = ""
+    managed_hooks_dir: str = ""
+    hooks: list[dict[str, Any]] = Field(default_factory=list)
+    eligible: bool | None = None
+    verbose: bool | None = None
+
+
+class PluginInspectEntry(BaseModel):
+    plugin: dict[str, Any] = Field(default_factory=dict)
+    shape: str | None = None
+    capabilityMode: str | None = None
+    capabilityCount: int = 0
+    capabilities: list[dict[str, Any]] = Field(default_factory=list)
+    typedHooks: list[dict[str, Any]] = Field(default_factory=list)
+    customHooks: list[dict[str, Any]] = Field(default_factory=list)
+    tools: list[dict[str, Any]] = Field(default_factory=list)
+    commands: list[str] = Field(default_factory=list)
+    cliCommands: list[str] = Field(default_factory=list)
+    services: list[str] = Field(default_factory=list)
+    gatewayMethods: list[str] = Field(default_factory=list)
+    mcpServers: list[dict[str, Any]] = Field(default_factory=list)
+    lspServers: list[dict[str, Any]] = Field(default_factory=list)
+    httpRouteCount: int = 0
+    bundleCapabilities: list[str] = Field(default_factory=list)
+
+
+class PluginsInspectResponse(BaseModel):
+    inspect: list[dict[str, Any]] = Field(default_factory=list)
+
+
+class AgentBindingItem(BaseModel):
+    agentId: str
+    match: dict[str, Any]
+    description: str
+
+
+class AgentsBindingsResponse(BaseModel):
+    bindings: list[AgentBindingItem]
+
+
+# ---------------------------------------------------------------------------
+# Routes — use typed model methods and return Pydantic models directly
+# ---------------------------------------------------------------------------
+
+@router.get("/status")
+async def api_openclaw_status(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> OpenClawStatus:
+    """Read `openclaw status --json` and return a typed model."""
+    try:
+        return service.status_model()
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/sessions")
+async def api_openclaw_sessions(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SessionsResponse:
+    """Read `openclaw sessions --json` and return a typed SessionsList."""
+    try:
+        result = service.list_sessions_model()
+        return SessionsResponse(sessions=result.sessions)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/sessions/{session_key:path}/history")
+async def api_openclaw_session_history(
+    session_key: str,
+    limit: int = Query(20, ge=1, le=200),
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SessionHistoryResponse:
+    """Read session history and return a typed SessionHistory."""
+    try:
+        result = service.get_session_history_model(session_key, limit=limit)
+        return SessionHistoryResponse(
+            session_key=result.session_key,
+            session_id=result.session_id,
+            events=result.events,
+            history=result.events,  # alias for compat
+            raw_text=result.raw_text,
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/sessions/{session_key:path}")
+async def api_openclaw_session_detail(
+    session_key: str,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SessionDetailResponse:
+    """Resolve a single session and return it as a typed model."""
+    try:
+        session = service.get_session_model(session_key)
+        return SessionDetailResponse(session=session)
+    except KeyError as exc:
+        raise HTTPException(status_code=404, detail=f"session '{session_key}' not found") from exc
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/cron")
+async def api_openclaw_cron(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> CronResponse:
+    """Read `openclaw cron list --json` and return a typed CronList."""
+    try:
+        result = service.list_cron_jobs_model()
+        return CronResponse(cron=list(result.cron), jobs=list(result.jobs))
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/approvals")
+async def api_openclaw_approvals(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ApprovalsResponse:
+    """Read `openclaw approvals get --json` and return a typed ApprovalsList."""
+    try:
+        result = service.list_approvals_model()
+        return ApprovalsResponse(
+            approvals=list(result.approvals),
+            pending=list(result.pending),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/agents")
+async def api_openclaw_agents(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentsResponse:
+    """Read `openclaw agents list --json` and return a typed AgentsList."""
+    try:
+        result = service.list_agents_model()
+        return AgentsResponse(agents=list(result.agents))
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/agents/presence")
+async def api_openclaw_agents_presence(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> dict[str, Any]:
+    """Read runtime session presence for all agents from session files."""
+    result = service.agents_presence()
+    return result
+
+
+# ---------------------------------------------------------------------------
+# Write agents routes
+# ---------------------------------------------------------------------------
+
+
+class AgentAddResponse(BaseModel):
+    agentId: str
+    name: str
+    workspace: str
+    agentDir: str
+    model: str | None = None
+    bindings: dict[str, Any] = Field(default_factory=dict)
+
+
+class AgentDeleteResponse(BaseModel):
+    agentId: str
+    workspace: str
+    agentDir: str
+    sessionsDir: str
+    removedBindings: list[str] = Field(default_factory=list)
+    removedAllow: list[str] = Field(default_factory=list)
+
+
+class AgentBindResponse(BaseModel):
+    agentId: str
+    added: list[str] = Field(default_factory=list)
+    updated: list[str] = Field(default_factory=list)
+    skipped: list[str] = Field(default_factory=list)
+    conflicts: list[str] = Field(default_factory=list)
+
+
+class AgentUnbindResponse(BaseModel):
+    agentId: str
+    removed: list[str] = Field(default_factory=list)
+    missing: list[str] = Field(default_factory=list)
+    conflicts: list[str] = Field(default_factory=list)
+
+
+class AgentIdentityResponse(BaseModel):
+    agentId: str
+    identity: dict[str, Any] = Field(default_factory=dict)
+    workspace: str | None = None
+    identityFile: str | None = None
+
+
+@router.post("/agents/add")
+async def api_openclaw_agents_add(
+    name: str,
+    *,
+    workspace: str | None = None,
+    model: str | None = None,
+    agent_dir: str | None = None,
+    bind: list[str] | None = None,
+    non_interactive: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentAddResponse:
+    """Run `openclaw agents add <name>` and return JSON result."""
+    try:
+        result = service.agents_add(
+            name,
+            workspace=workspace,
+            model=model,
+            agent_dir=agent_dir,
+            bind=bind,
+            non_interactive=non_interactive,
+        )
+        return AgentAddResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.post("/agents/delete/{id}")
+async def api_openclaw_agents_delete(
+    id: str,
+    force: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentDeleteResponse:
+    """Run `openclaw agents delete <id> [--force]` and return JSON result."""
+    try:
+        result = service.agents_delete(id, force=force)
+        return AgentDeleteResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.post("/agents/bind")
+async def api_openclaw_agents_bind(
+    *,
+    agent: str | None = None,
+    bind: list[str] | None = None,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentBindResponse:
+    """Run `openclaw agents bind [--agent <id>] [--bind <spec>]` and return JSON result."""
+    try:
+        result = service.agents_bind(agent=agent, bind=bind)
+        return AgentBindResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.post("/agents/unbind")
+async def api_openclaw_agents_unbind(
+    *,
+    agent: str | None = None,
+    bind: list[str] | None = None,
+    all: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentUnbindResponse:
+    """Run `openclaw agents unbind [--agent <id>] [--bind <spec>] [--all]` and return JSON result."""
+    try:
+        result = service.agents_unbind(agent=agent, bind=bind, all=all)
+        return AgentUnbindResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.post("/agents/set-identity")
+async def api_openclaw_agents_set_identity(
+    *,
+    agent: str | None = None,
+    workspace: str | None = None,
+    identity_file: str | None = None,
+    name: str | None = None,
+    emoji: str | None = None,
+    theme: str | None = None,
+    avatar: str | None = None,
+    from_identity: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentIdentityResponse:
+    """Run `openclaw agents set-identity` and return JSON result."""
+    try:
+        result = service.agents_set_identity(
+            agent=agent,
+            workspace=workspace,
+            identity_file=identity_file,
+            name=name,
+            emoji=emoji,
+            theme=theme,
+            avatar=avatar,
+            from_identity=from_identity,
+        )
+        return AgentIdentityResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/skills")
+async def api_openclaw_skills(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SkillsResponse:
+    """Read `openclaw skills list --json` and return a typed SkillStatusReport."""
+    try:
+        result = service.list_skills_model()
+        return SkillsResponse(
+            workspace_dir=result.workspace_dir,
+            managed_skills_dir=result.managed_skills_dir,
+            skills=list(result.skills),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/models")
+async def api_openclaw_models(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ModelsResponse:
+    """Read `openclaw models list --json` and return a typed ModelsList."""
+    try:
+        result = service.list_models_model()
+        return ModelsResponse(models=list(result.models))
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/hooks")
+async def api_openclaw_hooks(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> HooksResponse:
+    try:
+        result = service.list_hooks_model()
+        return HooksResponse(
+            workspace_dir=result.workspace_dir,
+            managed_hooks_dir=result.managed_hooks_dir,
+            hooks=list(result.hooks),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/plugins")
+async def api_openclaw_plugins(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> PluginsResponse:
+    try:
+        result = service.list_plugins_model()
+        return PluginsResponse(
+            workspace_dir=result.workspace_dir,
+            plugins=list(result.plugins),
+            diagnostics=list(result.diagnostics),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/secrets-audit")
+async def api_openclaw_secrets_audit(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SecretsAuditResponse:
+    try:
+        result = service.secrets_audit_model()
+        return SecretsAuditResponse(
+            version=result.version,
+            status=result.status,
+            findings=list(result.findings),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/security-audit")
+async def api_openclaw_security_audit(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SecurityAuditResponse2:
+    try:
+        result = service.security_audit_model()
+        return SecurityAuditResponse2(
+            report=result.report.model_dump() if result.report else None,
+            secret_diagnostics=list(result.secret_diagnostics),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/daemon-status")
+async def api_openclaw_daemon_status(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> DaemonStatusResponse:
+    try:
+        result = service.daemon_status_model()
+        return DaemonStatusResponse(
+            service=result.service.model_dump() if result.service else None,
+            port=result.port.model_dump() if result.port else None,
+            rpc=result.rpc.model_dump() if result.rpc else None,
+            health=result.health.model_dump() if result.health else None,
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/pairing")
+async def api_openclaw_pairing(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> PairingListResponse2:
+    try:
+        result = service.pairing_list_model()
+        return PairingListResponse2(
+            channel=result.channel,
+            requests=list(result.requests),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/qr")
+async def api_openclaw_qr(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> QrCodeResponse2:
+    try:
+        result = service.qr_code_model()
+        return QrCodeResponse2(
+            setup_code=result.setup_code,
+            gateway_url=result.gateway_url,
+            auth=result.auth,
+            url_source=result.url_source,
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/update-status")
+async def api_openclaw_update_status(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> UpdateStatusResponse2:
+    try:
+        result = service.update_status_model()
+        return UpdateStatusResponse2(
+            update=result.update.model_dump() if result.update else None,
+            channel=result.channel.model_dump() if result.channel else None,
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/models-aliases")
+async def api_openclaw_models_aliases(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ModelAliasesResponse:
+    try:
+        result = service.list_model_aliases_model()
+        return ModelAliasesResponse(aliases=result.aliases)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/models-fallbacks")
+async def api_openclaw_models_fallbacks(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ModelFallbacksResponse:
+    try:
+        result = service.list_model_fallbacks_model()
+        return ModelFallbacksResponse(
+            key=result.key,
+            label=result.label,
+            items=list(result.items),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/models-image-fallbacks")
+async def api_openclaw_models_image_fallbacks(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ModelFallbacksResponse:
+    try:
+        result = service.list_model_image_fallbacks_model()
+        return ModelFallbacksResponse(
+            key=result.key,
+            label=result.label,
+            items=list(result.items),
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/skill-update")
+async def api_openclaw_skill_update(
+    slug: str | None = None,
+    all: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> SkillUpdateResponse:
+    try:
+        result = service.skill_update_model(slug=slug, all=all)
+        return SkillUpdateResponse(
+            ok=result.ok,
+            slug=result.slug,
+            version=result.version,
+            error=result.error,
+        )
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/models-status")
+async def api_openclaw_models_status(
+    probe: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ModelsStatusResponse:
+    """Read `openclaw models status --json [--probe]` and return a typed dict."""
+    try:
+        result = service.models_status_model(probe=probe)
+        return ModelsStatusResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/channels-status")
+async def api_openclaw_channels_status(
+    probe: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ChannelsStatusResponse:
+    """Read `openclaw channels status --json [--probe]` and return a typed dict."""
+    try:
+        result = service.channels_status_model(probe=probe)
+        return ChannelsStatusResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/channels-list")
+async def api_openclaw_channels_list(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> ChannelsListResponse:
+    """Read `openclaw channels list --json` and return a typed dict."""
+    try:
+        result = service.channels_list_model()
+        return ChannelsListResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/hooks/info/{name}")
+async def api_openclaw_hook_info(
+    name: str,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> HookInfoResponse:
+    """Read `openclaw hooks info <name> --json` and return a typed dict."""
+    try:
+        result = service.hook_info_model(name)
+        return HookInfoResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/hooks/check")
+async def api_openclaw_hooks_check(
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> HooksCheckResponse:
+    """Read `openclaw hooks check --json` and return a typed dict."""
+    try:
+        result = service.hooks_check_model()
+        return HooksCheckResponse.model_validate(result, strict=False)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/plugins-inspect")
+async def api_openclaw_plugins_inspect(
+    plugin_id: str | None = None,
+    all: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> PluginsInspectResponse:
+    """Read `openclaw plugins inspect --json [--all]` and return a typed dict."""
+    try:
+        result = service.plugins_inspect_model(plugin_id=plugin_id, all=all)
+        inspect = result if isinstance(result, list) else result.get("inspect", [])
+        return PluginsInspectResponse(inspect=inspect)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+class AgentBindingItem(BaseModel):
+    agentId: str
+    match: dict[str, Any]
+    description: str
+
+
+class AgentsBindingsResponse(BaseModel):
+    bindings: list[AgentBindingItem]
+
+
+@router.get("/agents-bindings")
+async def api_openclaw_agents_bindings(
+    agent: str | None = None,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> AgentsBindingsResponse:
+    """Read `openclaw agents bindings --json [--agent <id>]` and return bindings list."""
+    try:
+        result = service.agents_bindings_model(agent=agent)
+        bindings = result if isinstance(result, list) else []
+        return AgentsBindingsResponse(bindings=bindings)
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/gateway-status")
+async def api_openclaw_gateway_status(
+    url: str | None = None,
+    token: str | None = None,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> dict[str, Any]:
+    """Read `openclaw gateway status --json [--url <url>] [--token <token>]`. Returns full gateway probe result."""
+    try:
+        result = service.gateway_status(url=url, token=token)
+        return result
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+@router.get("/memory-status")
+async def api_openclaw_memory_status(
+    agent: str | None = None,
+    deep: bool = False,
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> list[dict[str, Any]]:
+    """Read `openclaw memory status --json [--agent <id>] [--deep]`. Returns array of per-agent memory status."""
+    try:
+        result = service.memory_status(agent=agent, deep=deep)
+        return result if isinstance(result, list) else []
+    except OpenClawCliError as exc:
+        _raise_cli_http_error(exc)
+
+
+class WorkspaceFilesResponse(BaseModel):
+    workspace: str
+    files: list[dict[str, Any]]
+    error: str | None = None
+
+
+@router.get("/workspace-files")
+async def api_openclaw_workspace_files(
+    workspace: str = Query(..., description="Path to the agent workspace directory"),
+    service: OpenClawCliService = Depends(get_openclaw_cli_service),
+) -> WorkspaceFilesResponse:
+    """List .md files in an OpenClaw agent workspace with their content previews."""
+    result = service.list_workspace_files(workspace)
+    return WorkspaceFilesResponse.model_validate(result, strict=False)

backend/api/runtime.py

@@ -8,6 +8,7 @@ import json
 import logging
 import os
 import signal
+import shutil
 import subprocess
 import sys
 from datetime import datetime
@@ -38,9 +39,10 @@ class RuntimeState:
     """
 
     _instance: Optional["RuntimeState"] = None
-    _lock: asyncio.Lock = asyncio.Lock()
+    _lock: "threading.Lock" = __import__("threading").Lock()
 
     def __new__(cls) -> "RuntimeState":
+        with cls._lock:
             if cls._instance is None:
                 cls._instance = super().__new__(cls)
                 cls._instance._initialized = False
@@ -165,6 +167,8 @@ class RuntimeEventsResponse(BaseModel):
 
 class LaunchConfig(BaseModel):
     """Configuration for launching a new trading task."""
+    launch_mode: str = Field(default="fresh", description="启动形式: fresh, restore")
+    restore_run_id: Optional[str] = Field(default=None, description="历史任务 run_id，用于恢复启动")
     tickers: List[str] = Field(default_factory=list, description="股票池")
     schedule_mode: str = Field(default="daily", description="调度模式: daily, interval")
     interval_minutes: int = Field(default=60, ge=1, description="间隔分钟数")
@@ -177,7 +181,6 @@ class LaunchConfig(BaseModel):
     start_date: Optional[str] = Field(default=None, description="回测开始日期 YYYY-MM-DD")
     end_date: Optional[str] = Field(default=None, description="回测结束日期 YYYY-MM-DD")
     poll_interval: int = Field(default=10, ge=1, le=300, description="市场数据轮询间隔(秒)")
-    enable_mock: bool = Field(default=False, description="是否启用模拟模式(使用模拟价格数据)")
 
 
 class LaunchResponse(BaseModel):
@@ -188,11 +191,30 @@ class LaunchResponse(BaseModel):
     message: str
 
 
+class RuntimeHistoryItem(BaseModel):
+    run_id: str
+    run_dir: str
+    updated_at: Optional[str] = None
+    total_trades: int = 0
+    total_asset_value: Optional[float] = None
+    bootstrap: Dict[str, Any] = Field(default_factory=dict)
+
+
+class RuntimeHistoryResponse(BaseModel):
+    runs: List[RuntimeHistoryItem]
+
+
 class StopResponse(BaseModel):
     status: str
     message: str
 
 
+class CleanupResponse(BaseModel):
+    status: str
+    kept: int
+    pruned_run_ids: List[str]
+
+
 class GatewayStatusResponse(BaseModel):
     is_running: bool
     port: int
@@ -207,6 +229,13 @@ class RuntimeConfigResponse(BaseModel):
     resolved: Dict[str, Any]
 
 
+class RuntimeLogResponse(BaseModel):
+    run_id: Optional[str] = None
+    is_running: bool
+    log_path: Optional[str] = None
+    content: str = ""
+
+
 class UpdateRuntimeConfigRequest(BaseModel):
     schedule_mode: Optional[str] = None
     interval_minutes: Optional[int] = Field(default=None, ge=1)
@@ -227,6 +256,128 @@ def _get_run_dir(run_id: str) -> Path:
     return PROJECT_ROOT / "runs" / run_id
 
 
+def _load_run_snapshot(run_id: str) -> Dict[str, Any]:
+    """Load a specific run snapshot by run_id."""
+    snapshot_path = _get_run_dir(run_id) / "state" / "runtime_state.json"
+    if not snapshot_path.exists():
+        raise HTTPException(status_code=404, detail=f"Run snapshot not found: {run_id}")
+    return json.loads(snapshot_path.read_text(encoding="utf-8"))
+
+
+def _copy_path_if_exists(src: Path, dst: Path) -> None:
+    if not src.exists():
+        return
+    if src.is_dir():
+        shutil.copytree(src, dst, dirs_exist_ok=True)
+    else:
+        dst.parent.mkdir(parents=True, exist_ok=True)
+        shutil.copy2(src, dst)
+
+
+def _restore_run_assets(source_run_id: str, target_run_dir: Path) -> None:
+    """Seed a fresh run directory from a historical run snapshot."""
+    source_run_dir = _get_run_dir(source_run_id)
+    if not source_run_dir.exists():
+        raise HTTPException(status_code=404, detail=f"Source run not found: {source_run_id}")
+
+    for relative in [
+        "team_dashboard",
+        "agents",
+        "skills",
+        "memory",
+        "state/server_state.json",
+        "state/runtime.db",
+        "state/research.db",
+    ]:
+        _copy_path_if_exists(source_run_dir / relative, target_run_dir / relative)
+
+
+def _list_runs(limit: int = 50) -> list[RuntimeHistoryItem]:
+    runs_root = PROJECT_ROOT / "runs"
+    if not runs_root.exists():
+        return []
+
+    items: list[RuntimeHistoryItem] = []
+    run_dirs = sorted(
+        [path for path in runs_root.iterdir() if path.is_dir()],
+        key=lambda path: path.stat().st_mtime,
+        reverse=True,
+    )
+
+    for run_dir in run_dirs[: max(1, int(limit))]:
+        run_id = run_dir.name
+        runtime_state_path = run_dir / "state" / "runtime_state.json"
+        summary_path = run_dir / "team_dashboard" / "summary.json"
+
+        bootstrap: Dict[str, Any] = {}
+        updated_at: Optional[str] = None
+        total_trades = 0
+        total_asset_value: Optional[float] = None
+
+        if runtime_state_path.exists():
+            try:
+                snapshot = json.loads(runtime_state_path.read_text(encoding="utf-8"))
+                context = snapshot.get("context") or {}
+                bootstrap = dict(context.get("bootstrap_values") or {})
+                updated_at = snapshot.get("events", [{}])[-1].get("timestamp") if snapshot.get("events") else None
+            except Exception:
+                bootstrap = {}
+
+        if summary_path.exists():
+            try:
+                summary = json.loads(summary_path.read_text(encoding="utf-8"))
+                total_trades = int(summary.get("totalTrades") or 0)
+                total_asset_value = float(summary.get("totalAssetValue")) if summary.get("totalAssetValue") is not None else None
+            except Exception:
+                total_trades = 0
+                total_asset_value = None
+
+        items.append(
+            RuntimeHistoryItem(
+                run_id=run_id,
+                run_dir=str(run_dir),
+                updated_at=updated_at,
+                total_trades=total_trades,
+                total_asset_value=total_asset_value,
+                bootstrap=bootstrap,
+            )
+        )
+
+    return items
+
+
+def _is_timestamped_run_dir(path: Path) -> bool:
+    try:
+        datetime.strptime(path.name, "%Y%m%d_%H%M%S")
+        return True
+    except ValueError:
+        return False
+
+
+def _prune_old_timestamped_runs(*, keep: int = 20, exclude_run_ids: Optional[set[str]] = None) -> list[str]:
+    """Prune old timestamped run directories, preserving the newest N and excluded ids."""
+    exclude = exclude_run_ids or set()
+    runs_root = PROJECT_ROOT / "runs"
+    if not runs_root.exists():
+        return []
+
+    candidates = sorted(
+        [
+            path
+            for path in runs_root.iterdir()
+            if path.is_dir() and _is_timestamped_run_dir(path) and path.name not in exclude
+        ],
+        key=lambda path: path.name,
+        reverse=True,
+    )
+
+    pruned: list[str] = []
+    for path in candidates[max(0, keep):]:
+        shutil.rmtree(path, ignore_errors=True)
+        pruned.append(path.name)
+    return pruned
+
+
 def _find_available_port(start_port: int = 8765, max_port: int = 9000) -> int:
     """Find an available port for Gateway."""
     import socket
@@ -238,11 +389,21 @@ def _find_available_port(start_port: int = 8765, max_port: int = 9000) -> int:
 
 
 def _is_gateway_running() -> bool:
-    """Check if Gateway process is running."""
+    """Check if Gateway process is running.
+
+    Checks both the internally-managed gateway process and falls back to
+    port availability (for externally-managed gateway processes).
+    """
     process = _runtime_state.gateway_process
-    if process is None:
+    if process is not None and process.poll() is None:
+        return True
+    # Fallback: check if the gateway port is in use (for externally started gateway)
+    import socket
+    try:
+        with socket.create_connection(("127.0.0.1", _runtime_state.gateway_port), timeout=1):
+            return True
+    except OSError:
         return False
-    return process.poll() is None
 
 
 def _stop_gateway() -> bool:
@@ -288,29 +449,29 @@ def _start_gateway_process(
         "--bootstrap", json.dumps(bootstrap)
     ]
 
-    # Start process
+    log_path = run_dir / "logs" / "gateway.log"
+    log_path.parent.mkdir(parents=True, exist_ok=True)
+
+    log_file = log_path.open("ab")
+    try:
         process = subprocess.Popen(
             cmd,
             env=env,
-        stdout=subprocess.PIPE,
+            stdout=log_file,
-        stderr=subprocess.PIPE,
+            stderr=subprocess.STDOUT,
             cwd=PROJECT_ROOT
         )
+    finally:
+        log_file.close()
 
     return process
 
 
 @router.get("/context", response_model=RunContextResponse)
 async def get_run_context() -> RunContextResponse:
-    """Return the most recent run context."""
+    """Return active runtime context, or latest persisted context when stopped."""
-    snapshot_path = PROJECT_ROOT.glob("runs/*/state/runtime_state.json")
+    snapshot = _get_active_runtime_snapshot() if _is_gateway_running() else _load_latest_runtime_snapshot()
-    snapshots = sorted(snapshot_path, key=lambda p: p.stat().st_mtime, reverse=True)
+    context = snapshot.get("context")
-
-    if not snapshots:
-        raise HTTPException(status_code=404, detail="No run context available")
-
-    latest = json.loads(snapshots[0].read_text(encoding="utf-8"))
-    context = latest.get("context")
     if context is None:
         raise HTTPException(status_code=404, detail="Run context is not ready")
 
@@ -323,15 +484,9 @@ async def get_run_context() -> RunContextResponse:
 
 @router.get("/agents", response_model=RuntimeAgentsResponse)
 async def get_runtime_agents() -> RuntimeAgentsResponse:
-    """Return agent states from the most recent run."""
+    """Return agent states from the active runtime, or latest persisted run."""
-    snapshot_path = PROJECT_ROOT.glob("runs/*/state/runtime_state.json")
+    snapshot = _get_active_runtime_snapshot() if _is_gateway_running() else _load_latest_runtime_snapshot()
-    snapshots = sorted(snapshot_path, key=lambda p: p.stat().st_mtime, reverse=True)
+    agents = snapshot.get("agents", [])
-
-    if not snapshots:
-        raise HTTPException(status_code=404, detail="No runtime state available")
-
-    latest = json.loads(snapshots[0].read_text(encoding="utf-8"))
-    agents = latest.get("agents", [])
 
     return RuntimeAgentsResponse(
         agents=[RuntimeAgentState(**a) for a in agents]
@@ -340,21 +495,21 @@ async def get_runtime_agents() -> RuntimeAgentsResponse:
 
 @router.get("/events", response_model=RuntimeEventsResponse)
 async def get_runtime_events() -> RuntimeEventsResponse:
-    """Return events from the most recent run."""
+    """Return events from the active runtime, or latest persisted run."""
-    snapshot_path = PROJECT_ROOT.glob("runs/*/state/runtime_state.json")
+    snapshot = _get_active_runtime_snapshot() if _is_gateway_running() else _load_latest_runtime_snapshot()
-    snapshots = sorted(snapshot_path, key=lambda p: p.stat().st_mtime, reverse=True)
+    events = snapshot.get("events", [])
-
-    if not snapshots:
-        raise HTTPException(status_code=404, detail="No runtime state available")
-
-    latest = json.loads(snapshots[0].read_text(encoding="utf-8"))
-    events = latest.get("events", [])
 
     return RuntimeEventsResponse(
         events=[RuntimeEvent(**e) for e in events]
     )
 
 
+@router.get("/history", response_model=RuntimeHistoryResponse)
+async def get_runtime_history(limit: int = 20) -> RuntimeHistoryResponse:
+    """List recent historical runs for restore/start selection."""
+    return RuntimeHistoryResponse(runs=_list_runs(limit=limit))
+
+
 @router.get("/gateway/status", response_model=GatewayStatusResponse)
 async def get_gateway_status() -> GatewayStatusResponse:
     """Get Gateway process status and port."""
@@ -362,15 +517,10 @@ async def get_gateway_status() -> GatewayStatusResponse:
     run_id = None
 
     if is_running:
-        # Try to find run_id from runtime state
-        snapshot_path = PROJECT_ROOT.glob("runs/*/state/runtime_state.json")
-        snapshots = sorted(snapshot_path, key=lambda p: p.stat().st_mtime, reverse=True)
-        if snapshots:
         try:
-                latest = json.loads(snapshots[0].read_text(encoding="utf-8"))
+            run_id = _get_active_runtime_context().get("config_name")
-                run_id = latest.get("context", {}).get("config_name")
         except Exception as e:
-                logger.warning(f"Failed to parse latest snapshot: {e}")
+            logger.warning(f"Failed to resolve active runtime context: {e}")
 
     return GatewayStatusResponse(
         is_running=is_running,
@@ -390,6 +540,26 @@ async def get_gateway_port(request: Request) -> Dict[str, Any]:
     }
 
 
+@router.get("/logs", response_model=RuntimeLogResponse)
+async def get_runtime_logs() -> RuntimeLogResponse:
+    """Return current runtime log tail, or the latest run log if runtime is stopped."""
+    try:
+        context = _get_active_runtime_context() if _is_gateway_running() else _get_runtime_context_from_latest_snapshot()
+    except HTTPException:
+        return RuntimeLogResponse(is_running=False, content="")
+
+    run_id = str(context.get("config_name") or "").strip() or None
+    log_path = _get_gateway_log_path_for_run(run_id) if run_id else None
+    content = _read_log_tail(log_path) if log_path else ""
+
+    return RuntimeLogResponse(
+        run_id=run_id,
+        is_running=_is_gateway_running(),
+        log_path=str(log_path) if log_path else None,
+        content=content,
+    )
+
+
 def _build_gateway_ws_url(request: Request, port: int) -> str:
     """Build a proxy-safe Gateway WebSocket URL."""
     forwarded_proto = request.headers.get("x-forwarded-proto", "").split(",")[0].strip()
@@ -416,10 +586,23 @@ def _load_latest_runtime_snapshot() -> Dict[str, Any]:
     return json.loads(snapshots[0].read_text(encoding="utf-8"))
 
 
-def _get_current_runtime_context() -> Dict[str, Any]:
+def _get_active_runtime_snapshot() -> Dict[str, Any]:
-    """Return the active runtime context from the latest snapshot."""
+    """Return the active runtime snapshot, preferring in-memory manager state."""
     if not _is_gateway_running():
         raise HTTPException(status_code=404, detail="No runtime is currently running")
+
+    manager = _runtime_state.runtime_manager
+    if manager is not None and hasattr(manager, "build_snapshot"):
+        snapshot = manager.build_snapshot()
+        context = snapshot.get("context") or {}
+        if context.get("config_name"):
+            return snapshot
+
+    return _load_latest_runtime_snapshot()
+
+
+def _get_runtime_context_from_latest_snapshot() -> Dict[str, Any]:
+    """Return the latest persisted runtime context regardless of active process state."""
     latest = _load_latest_runtime_snapshot()
     context = latest.get("context") or {}
     if not context.get("config_name"):
@@ -427,6 +610,35 @@ def _get_current_runtime_context() -> Dict[str, Any]:
     return context
 
 
+def _get_gateway_log_path_for_run(run_id: str) -> Path:
+    return _get_run_dir(run_id) / "logs" / "gateway.log"
+
+
+def _read_log_tail(path: Path, max_chars: int = 120_000) -> str:
+    if not path.exists() or not path.is_file():
+        return ""
+    text = path.read_text(encoding="utf-8", errors="replace")
+    if len(text) <= max_chars:
+        return text
+    return text[-max_chars:]
+
+
+def _get_current_runtime_context() -> Dict[str, Any]:
+    """Return the active runtime context from the latest snapshot."""
+    if not _is_gateway_running():
+        raise HTTPException(status_code=404, detail="No runtime is currently running")
+    snapshot = _get_active_runtime_snapshot()
+    context = snapshot.get("context") or {}
+    if not context.get("config_name"):
+        raise HTTPException(status_code=404, detail="No runtime context available")
+    return context
+
+
+def _get_active_runtime_context() -> Dict[str, Any]:
+    """Return the active runtime context, preferring in-memory runtime manager state."""
+    return _get_current_runtime_context()
+
+
 def _resolve_runtime_response(run_id: str) -> RuntimeConfigResponse:
     """Build a normalized runtime config response for the active run."""
     context = _get_current_runtime_context()
@@ -517,12 +729,30 @@ async def start_runtime(
         _stop_gateway()
         await asyncio.sleep(1)  # Wait for port release
 
-    # 2. Generate run ID and directory
+    launch_mode = str(config.launch_mode or "fresh").strip().lower()
+    if launch_mode not in {"fresh", "restore"}:
+        raise HTTPException(status_code=400, detail="launch_mode must be 'fresh' or 'restore'")
+
+    # 2. Resolve run ID, directory, and bootstrap
+    if launch_mode == "restore":
+        restore_run_id = str(config.restore_run_id or "").strip()
+        if not restore_run_id:
+            raise HTTPException(status_code=400, detail="restore_run_id is required when launch_mode=restore")
+        snapshot = _load_run_snapshot(restore_run_id)
+        context = snapshot.get("context") or {}
+        if not context.get("config_name"):
+            raise HTTPException(status_code=404, detail=f"Run context not found: {restore_run_id}")
+        run_id = restore_run_id
+        run_dir = _get_run_dir(run_id)
+        bootstrap = dict(context.get("bootstrap_values") or {})
+        bootstrap["launch_mode"] = "restore"
+        bootstrap["restore_run_id"] = restore_run_id
+    else:
         run_id = _generate_run_id()
         run_dir = _get_run_dir(run_id)
-
-    # 3. Prepare bootstrap config
         bootstrap = {
+            "launch_mode": "fresh",
+            "restore_run_id": None,
             "tickers": config.tickers,
             "schedule_mode": config.schedule_mode,
             "interval_minutes": config.interval_minutes,
@@ -535,9 +765,16 @@ async def start_runtime(
             "start_date": config.start_date,
             "end_date": config.end_date,
             "poll_interval": config.poll_interval,
-        "enable_mock": config.enable_mock,
         }
 
+    retention_keep = max(1, int(os.getenv("RUNS_RETENTION_COUNT", "20") or "20"))
+    pruned_run_ids = _prune_old_timestamped_runs(
+        keep=retention_keep,
+        exclude_run_ids={run_id},
+    )
+    if pruned_run_ids:
+        logger.info("Pruned old run directories: %s", ", ".join(pruned_run_ids))
+
     # 4. Create runtime manager
     manager = TradingRuntimeManager(
         config_name=run_id,
@@ -567,11 +804,12 @@ async def start_runtime(
         await asyncio.sleep(2)
 
         if not _is_gateway_running():
-            stdout, stderr = process.communicate(timeout=1)
             _runtime_state.gateway_process = None
+            log_path = _get_gateway_log_path_for_run(run_id)
+            log_tail = _read_log_tail(log_path, max_chars=4000)
             raise HTTPException(
                 status_code=500,
-                detail=f"Gateway failed to start: {stderr.decode() if stderr else 'Unknown error'}"
+                detail=f"Gateway failed to start: {log_tail or 'Unknown error'}"
             )
 
     except Exception as e:
@@ -637,6 +875,25 @@ async def stop_runtime(force: bool = True) -> StopResponse:
     )
 
 
+@router.post("/cleanup", response_model=CleanupResponse)
+async def cleanup_old_runs(keep: int = 20) -> CleanupResponse:
+    """Prune old timestamped run directories while preserving named runs."""
+    keep_count = max(1, int(keep))
+    exclude: set[str] = set()
+
+    if _is_gateway_running():
+        try:
+            active_context = _get_active_runtime_context()
+            active_run_id = str(active_context.get("config_name") or "").strip()
+            if active_run_id:
+                exclude.add(active_run_id)
+        except HTTPException:
+            pass
+
+    pruned = _prune_old_timestamped_runs(keep=keep_count, exclude_run_ids=exclude)
+    return CleanupResponse(status="ok", kept=keep_count, pruned_run_ids=pruned)
+
+
 @router.post("/restart")
 async def restart_runtime(
     config: LaunchConfig,
@@ -663,15 +920,7 @@ async def get_current_runtime():
     if not _is_gateway_running():
         raise HTTPException(status_code=404, detail="No runtime is currently running")
 
-    # Find latest runtime state
+    context = _get_active_runtime_context()
-    snapshot_path = PROJECT_ROOT.glob("runs/*/state/runtime_state.json")
-    snapshots = sorted(snapshot_path, key=lambda p: p.stat().st_mtime, reverse=True)
-
-    if not snapshots:
-        raise HTTPException(status_code=404, detail="No runtime information available")
-
-    latest = json.loads(snapshots[0].read_text(encoding="utf-8"))
-    context = latest.get("context", {})
 
     return {
         "run_id": context.get("config_name"),

backend/apps/__init__.py

@@ -5,10 +5,13 @@ from .agent_service import app as agent_app
 from .agent_service import create_app as create_agent_app
 from .news_service import app as news_app
 from .news_service import create_app as create_news_app
+from .openclaw_service import app as openclaw_app
+from .openclaw_service import create_app as create_openclaw_app
 from .runtime_service import app as runtime_app
 from .runtime_service import create_app as create_runtime_app
 from .trading_service import app as trading_app
 from .trading_service import create_app as create_trading_app
+from .cors import add_cors_middleware, get_cors_origins
 
 app = agent_app
 create_app = create_agent_app
@@ -20,8 +23,12 @@ __all__ = [
     "create_agent_app",
     "news_app",
     "create_news_app",
+    "openclaw_app",
+    "create_openclaw_app",
     "runtime_app",
     "create_runtime_app",
     "trading_app",
     "create_trading_app",
+    "add_cors_middleware",
+    "get_cors_origins",
 ]

backend/apps/agent_service.py

@@ -8,7 +8,8 @@ from pathlib import Path
 from typing import AsyncGenerator
 
 from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
+
+from backend.apps.cors import add_cors_middleware
 
 from backend.api import agents_router, guard_router, workspaces_router
 from backend.agents import AgentFactory, WorkspaceManager, get_registry
@@ -32,28 +33,22 @@ def create_app(project_root: Path | None = None) -> FastAPI:
         agent_factory.workspaces_root.mkdir(parents=True, exist_ok=True)
 
         registry = get_registry()
-        print("✓ EvoTraders API started")
+        print("✓ 大时代 API started")
         print(f"  - Workspaces root: {agent_factory.workspaces_root}")
         print(f"  - Registered agents: {registry.get_agent_count()}")
 
         yield
 
-        print("✓ EvoTraders API shutting down")
+        print("✓ 大时代 API shutting down")
 
     app = FastAPI(
-        title="EvoTraders Agent Service",
+        title="大时代 Agent Service",
-        description="REST API for the EvoTraders multi-agent control plane",
+        description="REST API for the 大时代 multi-agent control plane",
         version="0.1.0",
         lifespan=lifespan,
     )
 
-    app.add_middleware(
+    add_cors_middleware(app)
-        CORSMiddleware,
-        allow_origins=["*"],
-        allow_credentials=True,
-        allow_methods=["*"],
-        allow_headers=["*"],
-    )
 
     @app.get("/health")
     async def health_check() -> dict[str, object]:

backend/apps/cors.py

@@ -0,0 +1,30 @@
+# -*- coding: utf-8 -*-
+"""Shared CORS configuration for all microservice apps."""
+
+import os
+from typing import Sequence
+
+from fastapi.middleware.cors import CORSMiddleware
+
+
+def get_cors_origins() -> Sequence[str]:
+    """Get allowed CORS origins from environment variable.
+
+    Defaults to ["*"] for backward compatibility.
+    Set CORS_ALLOWED_ORIGINS env var (comma-separated) in production.
+    """
+    origins = os.getenv("CORS_ALLOWED_ORIGINS", "").strip()
+    if not origins:
+        return ["*"]
+    return [o.strip() for o in origins.split(",") if o.strip()]
+
+
+def add_cors_middleware(app: "FastAPI") -> None:
+    """Add CORS middleware to app with environment-configured origins."""
+    app.add_middleware(
+        CORSMiddleware,
+        allow_origins=get_cors_origins(),
+        allow_credentials=True,
+        allow_methods=["*"],
+        allow_headers=["*"],
+    )

backend/apps/news_service.py

@@ -6,32 +6,26 @@ from __future__ import annotations
 from typing import Any
 
 from fastapi import Depends, FastAPI, Query
-from fastapi.middleware.cors import CORSMiddleware
+from backend.apps.cors import add_cors_middleware
 
 from backend.data.market_store import MarketStore
 from backend.domains import news as news_domain
 
 
 def get_market_store() -> MarketStore:
-    """Create a market store dependency."""
+    """Get the MarketStore singleton dependency."""
-    return MarketStore()
+    return MarketStore.get_instance()
 
 
 def create_app() -> FastAPI:
     """Create the news/explain service app."""
     app = FastAPI(
-        title="EvoTraders News Service",
+        title="大时代 News Service",
         description="Read-only news enrichment and explain service surface extracted from the monolith",
         version="0.1.0",
     )
 
-    app.add_middleware(
+    add_cors_middleware(app)
-        CORSMiddleware,
-        allow_origins=["*"],
-        allow_credentials=True,
-        allow_methods=["*"],
-        allow_headers=["*"],
-    )
 
     @app.get("/health")
     async def health_check() -> dict[str, str]:
@@ -51,6 +45,7 @@ def create_app() -> FastAPI:
             start_date=start_date,
             end_date=end_date,
             limit=limit,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/news-for-date")
@@ -65,6 +60,7 @@ def create_app() -> FastAPI:
             ticker=ticker,
             date=date,
             limit=limit,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/news-timeline")
@@ -79,6 +75,7 @@ def create_app() -> FastAPI:
             ticker=ticker,
             start_date=start_date,
             end_date=end_date,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/categories")
@@ -95,6 +92,7 @@ def create_app() -> FastAPI:
             start_date=start_date,
             end_date=end_date,
             limit=limit,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/similar-days")
@@ -109,6 +107,7 @@ def create_app() -> FastAPI:
             ticker=ticker,
             date=date,
             n_similar=n_similar,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/stories/{ticker}")
@@ -121,6 +120,7 @@ def create_app() -> FastAPI:
             store,
             ticker=ticker,
             as_of_date=as_of_date,
+            refresh_if_stale=False,
         )
 
     @app.get("/api/range-explain")
@@ -139,6 +139,7 @@ def create_app() -> FastAPI:
             end_date=end_date,
             article_ids=article_ids,
             limit=limit,
+            refresh_if_stale=False,
         )
 
     return app

backend/apps/openclaw_service.py

@@ -0,0 +1,49 @@
+# -*- coding: utf-8 -*-
+"""Read-only OpenClaw CLI FastAPI surface."""
+
+from __future__ import annotations
+
+from fastapi import Depends, FastAPI
+
+from backend.api import openclaw_router
+from backend.apps.cors import add_cors_middleware
+from backend.api.openclaw import get_openclaw_cli_service
+
+
+def create_app() -> FastAPI:
+    """Create the OpenClaw service app."""
+    app = FastAPI(
+        title="大时代 OpenClaw Service",
+        description="Read-only OpenClaw CLI integration service surface",
+        version="0.1.0",
+    )
+
+    add_cors_middleware(app)
+
+    @app.get("/health")
+    async def health_check(
+        service=Depends(get_openclaw_cli_service),
+    ) -> dict[str, object]:
+        return service.health()
+
+    @app.get("/api/status")
+    async def api_status(
+        service=Depends(get_openclaw_cli_service),
+    ) -> dict[str, object]:
+        return {
+            "status": "operational",
+            "service": "openclaw-service",
+            "openclaw": service.health(),
+        }
+
+    app.include_router(openclaw_router)
+    return app
+
+
+app = create_app()
+
+
+if __name__ == "__main__":
+    import uvicorn
+
+    uvicorn.run(app, host="0.0.0.0", port=8004)

backend/apps/runtime_service.py

@@ -4,27 +4,21 @@
 from __future__ import annotations
 
 from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
 
 from backend.api import runtime_router
 from backend.api.runtime import get_runtime_state
+from backend.apps.cors import add_cors_middleware
 
 
 def create_app() -> FastAPI:
     """Create the runtime service app."""
     app = FastAPI(
-        title="EvoTraders Runtime Service",
+        title="大时代 Runtime Service",
         description="Runtime lifecycle and gateway service surface extracted from the monolith",
         version="0.1.0",
     )
 
-    app.add_middleware(
+    add_cors_middleware(app)
-        CORSMiddleware,
-        allow_origins=["*"],
-        allow_credentials=True,
-        allow_methods=["*"],
-        allow_headers=["*"],
-    )
 
     @app.get("/health")
     async def health_check() -> dict[str, object]:

backend/apps/trading_service.py

@@ -6,7 +6,7 @@ from __future__ import annotations
 from typing import Any
 
 from fastapi import FastAPI, Query
-from fastapi.middleware.cors import CORSMiddleware
+from backend.apps.cors import add_cors_middleware
 
 from backend.domains import trading as trading_domain
 from shared.schema import (
@@ -21,18 +21,12 @@ from shared.schema import (
 def create_app() -> FastAPI:
     """Create the trading data service app."""
     app = FastAPI(
-        title="EvoTraders Trading Service",
+        title="大时代 Trading Service",
         description="Read-only trading data service surface extracted from the monolith",
         version="0.1.0",
     )
 
-    app.add_middleware(
+    add_cors_middleware(app)
-        CORSMiddleware,
-        allow_origins=["*"],
-        allow_credentials=True,
-        allow_methods=["*"],
-        allow_headers=["*"],
-    )
 
     @app.get("/health")
     async def health_check() -> dict[str, str]:

backend/cli.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 """
-EvoTraders CLI - Command-line interface for the EvoTraders trading system.
+大时代 CLI - Command-line interface for the 大时代 trading system.
 
 This module provides easy-to-use commands for running backtest, live trading,
 and frontend development server.
@@ -44,7 +44,7 @@ from backend.enrich.news_enricher import enrich_symbols
 
 app = typer.Typer(
     name="evotraders",
-    help="EvoTraders: A self-evolving multi-agent trading system",
+    help="大时代：自进化多智能体交易系统",
     add_completion=False,
 )
 ingest_app = typer.Typer(help="Ingest Polygon market data into the research warehouse.")
@@ -919,7 +919,7 @@ def backtest(
     """
     console.print(
         Panel.fit(
-            "[bold cyan]EvoTraders Backtest Mode[/bold cyan]",
+            "[bold cyan]大时代 Backtest Mode[/bold cyan]",
             border_style="cyan",
         ),
     )
@@ -1019,11 +1019,6 @@ def backtest(
 
 @app.command()
 def live(
-    mock: bool = typer.Option(
-        False,
-        "--mock",
-        help="Use mock mode with simulated prices (for testing)",
-    ),
     config_name: str = typer.Option(
         "live",
         "--config-name",
@@ -1078,7 +1073,6 @@ def live(
 
     Example:
         evotraders live                    # Run immediately (default)
-        evotraders live --mock             # Mock mode
         evotraders live -t 22:30           # Run at 22:30 local time daily
         evotraders live --schedule-mode intraday --interval-minutes 60
         evotraders live --trigger-time now # Run immediately
@@ -1086,16 +1080,14 @@ def live(
     """
     schedule_mode = str(_normalize_typer_value(schedule_mode, "daily"))
     interval_minutes = int(_normalize_typer_value(interval_minutes, 60))
-    mode_name = "MOCK" if mock else "LIVE"
     console.print(
         Panel.fit(
-            f"[bold cyan]EvoTraders {mode_name} Mode[/bold cyan]",
+            "[bold cyan]大时代 LIVE Mode[/bold cyan]",
             border_style="cyan",
         ),
     )
 
     # Check for required API key in live mode
-    if not mock:
     env_file = get_project_root() / ".env"
     if not env_file.exists():
         console.print("\n[yellow]Warning: .env file not found[/yellow]")
@@ -1168,9 +1160,6 @@ def live(
 
     # Display configuration
     console.print("\n[bold]Configuration:[/bold]")
-    if mock:
-        console.print("   Mode: [yellow]MOCK[/yellow] (Simulated prices)")
-    else:
     console.print(
         "   Mode: [green]LIVE[/green] (Real-time prices via Finnhub)",
     )
@@ -1188,8 +1177,7 @@ def live(
     project_root = get_project_root()
     os.chdir(project_root)
 
-    # Data update (if not mock mode)
+    # Data update
-    if not mock:
     run_data_updater(project_root)
     auto_update_market_store(
         config_name,
@@ -1200,10 +1188,6 @@ def live(
         end_date=nyse_now.date().isoformat(),
         force=False,
     )
-    else:
-        console.print(
-            "\n[dim]Mock mode enabled - skipping data update[/dim]\n",
-        )
 
     # Build command using backend.main
     cmd = [
@@ -1229,8 +1213,6 @@ def live(
         str(interval_minutes),
     ]
 
-    if mock:
-        cmd.append("--mock")
     if enable_memory:
         cmd.append("--enable-memory")
 
@@ -1269,7 +1251,7 @@ def frontend(
     """
     console.print(
         Panel.fit(
-            "[bold cyan]EvoTraders Frontend[/bold cyan]",
+            "[bold cyan]大时代 Frontend[/bold cyan]",
             border_style="cyan",
         ),
     )
@@ -1337,16 +1319,16 @@ def frontend(
 
 @app.command()
 def version():
-    """Show the version of EvoTraders."""
+    """Show the version of 大时代."""
     console.print(
-        "\n[bold cyan]EvoTraders[/bold cyan] version [green]0.1.0[/green]\n",
+        "\n[bold cyan]大时代[/bold cyan] version [green]0.1.0[/green]\n",
     )
 
 
 @app.callback()
 def main():
     """
-    EvoTraders: A self-evolving multi-agent trading system
+    大时代：自进化多智能体交易系统
 
     Use 'evotraders --help' to see available commands.
     """

backend/config/bootstrap_config.py

@@ -4,6 +4,22 @@
 from dataclasses import dataclass, field
 from pathlib import Path
 from typing import Any, Dict
+
+
+DEFAULT_TICKERS = [
+    "AAPL",
+    "MSFT",
+    "GOOGL",
+    "AMZN",
+    "NVDA",
+    "META",
+    "TSLA",
+    "AMD",
+    "NFLX",
+    "AVGO",
+    "PLTR",
+    "COIN",
+]
 import re
 
 import yaml
@@ -127,7 +143,7 @@ def resolve_runtime_config(
     bootstrap = get_bootstrap_config_for_run(project_root, config_name)
     return {
         "tickers": bootstrap.get("tickers")
-        or get_env_list("TICKERS", ["AAPL", "MSFT"]),
+        or get_env_list("TICKERS", DEFAULT_TICKERS),
         "initial_cash": float(
             bootstrap.get(
                 "initial_cash",

backend/config/data_config.py

@@ -76,27 +76,19 @@ def _resolve_config() -> DataSourceConfig:
     """
     Resolve data source configuration based on available API keys.
 
-    Priority:
+    The effective source should always match the first item in the resolved
-    1. FINNHUB_API_KEY (if set)
+    ordered source list.
-    2. FINANCIAL_DATASETS_API_KEY (if set)
-    3. Raises error if neither is available
     """
     sources = _ordered_sources()
-    if "finnhub" in sources:
+    source = sources[0] if sources else "local_csv"
-        return DataSourceConfig(
+
-            source="finnhub",
+    api_key = ""
-            api_key=os.getenv("FINNHUB_API_KEY", "").strip(),
+    if source == "finnhub":
-            sources=sources,
+        api_key = os.getenv("FINNHUB_API_KEY", "").strip()
-        )
+    elif source == "financial_datasets":
-    if "financial_datasets" in sources:
+        api_key = os.getenv("FINANCIAL_DATASETS_API_KEY", "").strip()
-        return DataSourceConfig(
+
-            source="financial_datasets",
+    return DataSourceConfig(source=source, api_key=api_key, sources=sources)
-            api_key=os.getenv("FINANCIAL_DATASETS_API_KEY", "").strip(),
-            sources=sources,
-        )
-    if "yfinance" in sources:
-        return DataSourceConfig(source="yfinance", api_key="", sources=sources)
-    return DataSourceConfig(source="local_csv", api_key="", sources=sources)
 
 
 def get_config() -> DataSourceConfig:

backend/core/pipeline.py

@@ -18,7 +18,6 @@ from agentscope.message import Msg
 from agentscope.pipeline import MsgHub
 
 from backend.utils.settlement import SettlementCoordinator
-from backend.utils.terminal_dashboard import get_dashboard
 from backend.core.state_sync import StateSync
 from backend.utils.trade_executor import PortfolioTradeExecutor
 from backend.runtime.manager import TradingRuntimeManager
@@ -48,12 +47,8 @@ except ImportError:
 logger = logging.getLogger(__name__)
 
 
-def _log(msg: str):
+def _log(msg: str) -> None:
-    """Log to dashboard if available, otherwise to logger"""
+    """Helper function for pipeline logging."""
-    dashboard = get_dashboard()
-    if dashboard.live:
-        dashboard.log(msg)
-    else:
     logger.info(msg)
 
 
@@ -71,7 +66,7 @@ class TradingPipeline:
 
     Real-time updates via StateSync after each agent completes.
 
-    Supports both legacy agent lists and new workspace-based agent loading.
+    Supports both legacy agent lists and run-scoped agent loading.
     """
 
     def __init__(
@@ -1625,14 +1620,13 @@ class TradingPipeline:
         project_root = Path(__file__).resolve().parents[2]
         personas = get_prompt_loader().load_yaml_config("analyst", "personas")
         persona = personas.get(analyst_type, {})
-        WorkspaceManager(project_root=project_root).ensure_agent_assets(
+        workspace_manager = WorkspaceManager(project_root=project_root)
+        workspace_manager.ensure_agent_assets(
             config_name=config_name,
             agent_id=agent_id,
-            role_seed=persona.get("description", "").strip(),
+            file_contents=workspace_manager.build_default_agent_files(
-            style_seed="\n".join(f"- {item}" for item in persona.get("focus", [])),
+                agent_id=agent_id,
-            policy_seed=(
+                persona=persona,
-                "State a clear signal, confidence, and the conditions "
-                "that would invalidate the thesis."
             ),
         )
 

backend/core/pipeline_runner.py

@@ -232,7 +232,7 @@ async def run_pipeline(
 
     try:
         # Extract config values
-        tickers = bootstrap.get("tickers", ["AAPL", "MSFT"])
+        tickers = bootstrap.get("tickers", ["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA", "AMD", "NFLX", "AVGO", "PLTR", "COIN"])
         initial_cash = float(bootstrap.get("initial_cash", 100000.0))
         margin_requirement = float(bootstrap.get("margin_requirement", 0.0))
         max_comm_cycles = int(bootstrap.get("max_comm_cycles", 2))
@@ -244,10 +244,8 @@ async def run_pipeline(
         start_date = bootstrap.get("start_date")
         end_date = bootstrap.get("end_date")
         enable_memory = bootstrap.get("enable_memory", False)
-        enable_mock = bootstrap.get("enable_mock", False)
 
         is_backtest = mode == "backtest"
-        is_mock = enable_mock or mode == "mock" or (not is_backtest and os.getenv("MOCK_MODE", "false").lower() == "true")
 
         # ======================================================================
         # PHASE 0: Initialize runtime manager
@@ -266,10 +264,6 @@ async def run_pipeline(
 
         set_global_runtime_manager(runtime_manager)
 
-        # Register runtime manager with API
-        from backend.api.runtime import register_runtime_manager
-        register_runtime_manager(runtime_manager)
-
         # ======================================================================
         # PHASE 1 & 2: Create infrastructure services (Market, Storage)
         # These will be started by Gateway in the correct order
@@ -292,9 +286,8 @@ async def run_pipeline(
         market_service = MarketService(
             tickers=tickers,
             poll_interval=10,
-            mock_mode=is_mock and not is_backtest,
             backtest_mode=is_backtest,
-            api_key=os.getenv("FINNHUB_API_KEY") if not is_mock and not is_backtest else None,
+            api_key=os.getenv("FINNHUB_API_KEY") if not is_backtest else None,
             backtest_start_date=start_date if is_backtest else None,
             backtest_end_date=end_date if is_backtest else None,
         )
@@ -391,7 +384,6 @@ async def run_pipeline(
             scheduler_callback=scheduler_callback,
             config={
                 "mode": mode,
-                "mock_mode": is_mock,
                 "backtest_mode": is_backtest,
                 "tickers": tickers,
                 "config_name": run_id,

backend/core/state_sync.py

@@ -465,7 +465,6 @@ class StateSync:
 
         payload = {
             "server_mode": self._state.get("server_mode", "live"),
-            "is_mock_mode": self._state.get("is_mock_mode", False),
             "is_backtest": self._state.get("is_backtest", False),
             "tickers": self._state.get("tickers"),
             "runtime_config": self._state.get("runtime_config"),
@@ -488,12 +487,13 @@ class StateSync:
         }
 
         if include_dashboard:
+            dashboard_snapshot = self.storage.build_dashboard_snapshot_from_state(self._state)
             payload["dashboard"] = {
-                "summary": self.storage.load_file("summary"),
+                "summary": dashboard_snapshot.get("summary"),
-                "holdings": self.storage.load_file("holdings"),
+                "holdings": dashboard_snapshot.get("holdings"),
-                "stats": self.storage.load_file("stats"),
+                "stats": dashboard_snapshot.get("stats"),
-                "trades": self.storage.load_file("trades"),
+                "trades": dashboard_snapshot.get("trades"),
-                "leaderboard": self.storage.load_file("leaderboard"),
+                "leaderboard": dashboard_snapshot.get("leaderboard"),
             }
 
         return payload

backend/data/__init__.py

@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 from backend.data.historical_price_manager import HistoricalPriceManager
-from backend.data.mock_price_manager import MockPriceManager
 from backend.data.polling_price_manager import PollingPriceManager
 
-__all__ = ["MockPriceManager", "PollingPriceManager", "HistoricalPriceManager"]
+__all__ = ["PollingPriceManager", "HistoricalPriceManager"]

backend/data/market_ingest.py

@@ -8,6 +8,7 @@ from typing import Iterable
 
 from backend.data.market_store import MarketStore
 from backend.data.news_alignment import align_news_for_symbol
+from backend.data.provider_router import DataProviderRouter
 from backend.data.polygon_client import (
     fetch_news,
     fetch_ohlc,
@@ -24,6 +25,65 @@ def _default_start(years: int = 2) -> str:
     return (datetime.now(timezone.utc).date() - timedelta(days=years * 366)).isoformat()
 
 
+def _max_news_date(news_rows: Iterable[dict]) -> str | None:
+    dates = [
+        str(item.get("published_utc") or "").strip()[:10]
+        for item in news_rows
+        if str(item.get("published_utc") or "").strip()
+    ]
+    dates = [value for value in dates if value]
+    return max(dates) if dates else None
+
+
+def _effective_last_news_fetch(
+    market_store: MarketStore,
+    *,
+    ticker: str,
+    end_date: str,
+    watermark_value: str | None,
+) -> str | None:
+    """Clamp stale/future watermarks to the latest actually stored news date."""
+    raw = str(watermark_value or "").strip()[:10]
+    if not raw:
+        return None
+    if raw <= end_date:
+        return raw
+
+    latest_stored = market_store.get_latest_news_date(ticker)
+    if latest_stored and latest_stored <= end_date:
+        return latest_stored
+    return end_date
+
+
+def _normalize_provider_news_rows(ticker: str, news_items: Iterable[Any]) -> list[dict]:
+    rows: list[dict] = []
+    for item in news_items:
+        payload = item.model_dump() if hasattr(item, "model_dump") else dict(item or {})
+        related = payload.get("related")
+        if isinstance(related, str):
+            related_list = [value.strip().upper() for value in related.split(",") if value.strip()]
+        elif isinstance(related, list):
+            related_list = [str(value).strip().upper() for value in related if str(value).strip()]
+        else:
+            related_list = []
+        if ticker not in related_list:
+            related_list.append(ticker)
+        rows.append(
+            {
+                "title": payload.get("title"),
+                "description": payload.get("summary"),
+                "summary": payload.get("summary"),
+                "article_url": payload.get("url"),
+                "published_utc": payload.get("date"),
+                "publisher": payload.get("source"),
+                "tickers": related_list,
+                "category": payload.get("category"),
+                "raw_json": payload,
+            }
+        )
+    return rows
+
+
 def ingest_ticker_history(
     symbol: str,
     *,
@@ -50,7 +110,11 @@ def ingest_ticker_history(
     price_count = market_store.upsert_ohlc(ticker, ohlc_rows, source="polygon")
     news_count = market_store.upsert_news(ticker, news_rows, source="polygon")
     aligned_count = align_news_for_symbol(market_store, ticker)
-    market_store.update_fetch_watermark(symbol=ticker, price_date=end, news_date=end)
+    market_store.update_fetch_watermark(
+        symbol=ticker,
+        price_date=end,
+        news_date=_max_news_date(news_rows),
+    )
 
     return {
         "symbol": ticker,
@@ -78,9 +142,15 @@ def update_ticker_incremental(
         if watermarks.get("last_price_fetch")
         else _default_start()
     )
+    effective_last_news_fetch = _effective_last_news_fetch(
+        market_store,
+        ticker=ticker,
+        end_date=end,
+        watermark_value=watermarks.get("last_news_fetch"),
+    )
     start_news = (
-        (datetime.fromisoformat(watermarks["last_news_fetch"]) + timedelta(days=1)).date().isoformat()
+        (datetime.fromisoformat(effective_last_news_fetch) + timedelta(days=1)).date().isoformat()
-        if watermarks.get("last_news_fetch")
+        if effective_last_news_fetch
         else _default_start()
     )
 
@@ -100,7 +170,7 @@ def update_ticker_incremental(
     market_store.update_fetch_watermark(
         symbol=ticker,
         price_date=end if ohlc_rows or watermarks.get("last_price_fetch") else None,
-        news_date=end if news_rows or watermarks.get("last_news_fetch") else None,
+        news_date=_max_news_date(news_rows),
     )
 
     return {
@@ -114,6 +184,86 @@ def update_ticker_incremental(
     }
 
 
+def refresh_news_incremental(
+    symbol: str,
+    *,
+    end_date: str | None = None,
+    store: MarketStore | None = None,
+) -> dict:
+    """Incrementally fetch company news using the configured provider router."""
+    ticker = normalize_symbol(symbol)
+    market_store = store or MarketStore()
+    watermarks = market_store.get_ticker_watermarks(ticker)
+    end = end_date or _today_utc()
+    effective_last_news_fetch = _effective_last_news_fetch(
+        market_store,
+        ticker=ticker,
+        end_date=end,
+        watermark_value=watermarks.get("last_news_fetch"),
+    )
+    start_news = (
+        (datetime.fromisoformat(effective_last_news_fetch) + timedelta(days=1)).date().isoformat()
+        if effective_last_news_fetch
+        else _default_start()
+    )
+
+    if start_news > end:
+        return {
+            "symbol": ticker,
+            "start_news_date": start_news,
+            "end_date": end,
+            "news": 0,
+            "aligned": 0,
+        }
+
+    router = DataProviderRouter()
+    news_items, source = router.get_company_news(
+        ticker=ticker,
+        start_date=start_news,
+        end_date=end,
+        limit=1000,
+    )
+    news_rows = _normalize_provider_news_rows(ticker, news_items)
+    news_count = market_store.upsert_news(ticker, news_rows, source=source) if news_rows else 0
+    aligned_count = align_news_for_symbol(market_store, ticker)
+    market_store.update_fetch_watermark(
+        symbol=ticker,
+        news_date=_max_news_date(news_rows),
+    )
+
+    return {
+        "symbol": ticker,
+        "start_news_date": start_news,
+        "end_date": end,
+        "news": news_count,
+        "aligned": aligned_count,
+        "source": source,
+    }
+
+
+def refresh_news_for_symbols(
+    symbols: Iterable[str],
+    *,
+    end_date: str | None = None,
+    store: MarketStore | None = None,
+) -> list[dict]:
+    """Incrementally refresh company news for a list of tickers."""
+    market_store = store or MarketStore()
+    results = []
+    for symbol in symbols:
+        ticker = normalize_symbol(symbol)
+        if not ticker:
+            continue
+        results.append(
+            refresh_news_incremental(
+                ticker,
+                end_date=end_date,
+                store=market_store,
+            )
+        )
+    return results
+
+
 def ingest_symbols(
     symbols: Iterable[str],
     *,

backend/data/market_store.py

@@ -9,7 +9,7 @@ import os
 import sqlite3
 from datetime import datetime, timezone
 from pathlib import Path
-from typing import Any, Iterable
+from typing import Any, Iterable, Optional
 
 
 SCHEMA = """
@@ -147,12 +147,30 @@ def _utc_timestamp() -> str:
 
 
 class MarketStore:
-    """SQLite-backed market research warehouse."""
+    """SQLite-backed market research warehouse. Use get_instance() for the singleton."""
+
+    _instance: Optional["MarketStore"] = None
+
+    def __new__(cls, db_path: Path | None = None) -> "MarketStore":
+        if cls._instance is not None:
+            if db_path is None or cls._instance.db_path == Path(db_path or get_market_db_path()):
+                return cls._instance
+        instance = super().__new__(cls)
+        cls._instance = instance
+        return instance
 
     def __init__(self, db_path: Path | None = None):
+        if getattr(self, "_initialized", False):
+            return
         self.db_path = Path(db_path or get_market_db_path())
         self.db_path.parent.mkdir(parents=True, exist_ok=True)
         self._init_db()
+        self._initialized = True
+
+    @classmethod
+    def get_instance(cls, db_path: Path | None = None) -> "MarketStore":
+        """Get the MarketStore singleton instance."""
+        return cls(db_path)
 
     def _connect(self) -> sqlite3.Connection:
         conn = sqlite3.connect(self.db_path)
@@ -226,6 +244,20 @@ class MarketStore:
             "last_news_fetch": None,
         }
 
+    def get_latest_news_date(self, symbol: str) -> str | None:
+        """Return the latest stored published news date for one ticker."""
+        with self._connect() as conn:
+            row = conn.execute(
+                """
+                SELECT MAX(substr(nr.published_utc, 1, 10)) AS latest_date
+                FROM news_ticker nt
+                JOIN news_raw nr ON nr.id = nt.news_id
+                WHERE nt.symbol = ?
+                """,
+                (symbol,),
+            ).fetchone()
+        return str(row["latest_date"]).strip() if row and row["latest_date"] else None
+
     def upsert_ohlc(self, symbol: str, rows: Iterable[dict[str, Any]], *, source: str = "polygon") -> int:
         timestamp = _utc_timestamp()
         count = 0

backend/data/mock_price_manager.py

@@ -1,244 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Mock Price Manager - For testing during non-trading hours
-Generates virtual real-time price data
-"""
-import logging
-import os
-import random
-import threading
-import time
-from typing import Callable, Dict, List, Optional
-from backend.data.provider_utils import normalize_symbol
-
-logger = logging.getLogger(__name__)
-
-
-class MockPriceManager:
-    """Mock Price Manager - Generates virtual prices for testing"""
-
-    def __init__(self, poll_interval: int = 10, volatility: float = 0.5):
-        """
-        Args:
-            poll_interval: Price update interval in seconds
-            volatility: Price volatility percentage
-        """
-        if poll_interval is None:
-            poll_interval = int(os.getenv("MOCK_POLL_INTERVAL", "5"))
-        if volatility is None:
-            volatility = float(os.getenv("MOCK_VOLATILITY", "0.5"))
-
-        self.poll_interval = poll_interval
-        self.volatility = volatility
-
-        self.subscribed_symbols: List[str] = []
-        self.base_prices: Dict[str, float] = {}
-        self.open_prices: Dict[str, float] = {}
-        self.latest_prices: Dict[str, float] = {}
-        self.price_callbacks: List[Callable] = []
-
-        self.running = False
-        self._thread: Optional[threading.Thread] = None
-
-        self.default_base_prices = {
-            "AAPL": 237.50,
-            "MSFT": 425.30,
-            "GOOGL": 161.50,
-            "AMZN": 218.45,
-            "NVDA": 950.00,
-            "META": 573.22,
-            "TSLA": 342.15,
-            "AMD": 168.90,
-            "NFLX": 688.25,
-            "INTC": 42.18,
-            "COIN": 285.50,
-            "PLTR": 45.80,
-            "BABA": 88.30,
-            "DIS": 112.50,
-            "BKNG": 4850.00,
-        }
-
-        logger.info(
-            f"MockPriceManager initialized (interval: {self.poll_interval}s, "
-            f"volatility: {self.volatility}%)",
-        )
-
-    def subscribe(
-        self,
-        symbols: List[str],
-        base_prices: Dict[str, float] = None,
-    ):
-        """Subscribe to stock symbols"""
-        for symbol in symbols:
-            symbol = normalize_symbol(symbol)
-            if symbol not in self.subscribed_symbols:
-                self.subscribed_symbols.append(symbol)
-
-                if base_prices and symbol in base_prices:
-                    base_price = base_prices[symbol]
-                elif symbol in self.default_base_prices:
-                    base_price = self.default_base_prices[symbol]
-                else:
-                    base_price = random.uniform(50, 500)
-
-                self.base_prices[symbol] = base_price
-                self.open_prices[symbol] = base_price
-                self.latest_prices[symbol] = base_price
-
-                logger.info(
-                    f"Subscribed to mock price: {symbol} (base: ${base_price:.2f})",  # noqa: E501
-                )
-
-    def unsubscribe(self, symbols: List[str]):
-        """Unsubscribe from symbols"""
-        for symbol in symbols:
-            symbol = normalize_symbol(symbol)
-            if symbol in self.subscribed_symbols:
-                self.subscribed_symbols.remove(symbol)
-                self.base_prices.pop(symbol, None)
-                self.open_prices.pop(symbol, None)
-                self.latest_prices.pop(symbol, None)
-                logger.info(f"Unsubscribed: {symbol}")
-
-    def add_price_callback(self, callback: Callable):
-        """Add price update callback"""
-        self.price_callbacks.append(callback)
-
-    def _generate_price_update(self, symbol: str) -> float:
-        """Generate price update based on random walk"""
-        current_price = self.latest_prices.get(
-            symbol,
-            self.base_prices[symbol],
-        )
-
-        change_percent = random.uniform(-self.volatility, self.volatility)
-        new_price = current_price * (1 + change_percent / 100)
-
-        # 10% chance of larger movement
-        if random.random() < 0.1:
-            trend_factor = random.uniform(-2, 2)
-            new_price = new_price * (1 + trend_factor / 100)
-
-        # Limit intraday movement to +/-10%
-        open_price = self.open_prices[symbol]
-        max_price = open_price * 1.10
-        min_price = open_price * 0.90
-        new_price = max(min_price, min(max_price, new_price))
-
-        return new_price
-
-    def _update_prices(self):
-        """Update prices for all subscribed stocks"""
-        timestamp = int(time.time() * 1000)
-
-        for symbol in self.subscribed_symbols:
-            try:
-                new_price = self._generate_price_update(symbol)
-                self.latest_prices[symbol] = new_price
-
-                open_price = self.open_prices[symbol]
-                ret = ((new_price - open_price) / open_price) * 100
-
-                price_data = {
-                    "symbol": symbol,
-                    "price": new_price,
-                    "timestamp": timestamp,
-                    "volume": random.randint(1000000, 10000000),
-                    "open": open_price,
-                    "high": max(new_price, open_price),
-                    "low": min(new_price, open_price),
-                    "previous_close": open_price,
-                    "ret": ret,
-                }
-
-                for callback in self.price_callbacks:
-                    try:
-                        callback(price_data)
-                    except Exception as e:
-                        logger.error(
-                            f"Mock price callback error ({symbol}): {e}",
-                        )
-
-                logger.debug(
-                    f"Mock {symbol}: ${new_price:.2f} [ret: {ret:+.2f}%]",
-                )
-
-            except Exception as e:
-                logger.error(f"Failed to generate mock price ({symbol}): {e}")
-
-    def _polling_loop(self):
-        """Main polling loop"""
-        logger.info(
-            f"Mock price generation started (interval: {self.poll_interval}s)",
-        )
-
-        while self.running:
-            try:
-                start_time = time.time()
-                self._update_prices()
-
-                elapsed = time.time() - start_time
-                sleep_time = max(0, self.poll_interval - elapsed)
-                if sleep_time > 0:
-                    time.sleep(sleep_time)
-
-            except Exception as e:
-                logger.error(f"Mock polling loop error: {e}")
-                time.sleep(5)
-
-    def start(self):
-        """Start mock price generation"""
-        if self.running:
-            logger.warning("Mock price manager already running")
-            return
-
-        if not self.subscribed_symbols:
-            logger.warning("No stocks subscribed")
-            return
-
-        self.running = True
-        self._thread = threading.Thread(target=self._polling_loop, daemon=True)
-        self._thread.start()
-
-        logger.info(
-            f"Mock price manager started: {', '.join(self.subscribed_symbols)}",  # noqa: E501
-        )
-
-    def stop(self):
-        """Stop mock price generation"""
-        self.running = False
-        if self._thread:
-            self._thread.join(timeout=5)
-        logger.info("Mock price manager stopped")
-
-    def get_latest_price(self, symbol: str) -> Optional[float]:
-        """Get latest price for symbol"""
-        return self.latest_prices.get(symbol)
-
-    def get_all_latest_prices(self) -> Dict[str, float]:
-        """Get all latest prices"""
-        return self.latest_prices.copy()
-
-    def get_open_price(self, symbol: str) -> Optional[float]:
-        """Get open price for symbol"""
-        return self.open_prices.get(symbol)
-
-    def reset_open_prices(self):
-        """Reset open prices for new trading day"""
-        for symbol in self.subscribed_symbols:
-            last_close = self.latest_prices[symbol]
-            gap_percent = random.uniform(-1, 1)
-            new_open = last_close * (1 + gap_percent / 100)
-            self.open_prices[symbol] = new_open
-            self.latest_prices[symbol] = new_open
-        logger.info("Open prices reset")
-
-    def set_base_price(self, symbol: str, price: float):
-        """Manually set base price for testing"""
-        if symbol in self.subscribed_symbols:
-            self.base_prices[symbol] = price
-            self.open_prices[symbol] = price
-            self.latest_prices[symbol] = price
-            logger.info(f"{symbol} base price set to: ${price:.2f}")
-        else:
-            logger.warning(f"{symbol} not subscribed")

backend/data/polling_price_manager.py

@@ -15,6 +15,9 @@ from backend.data.provider_utils import normalize_symbol
 logger = logging.getLogger(__name__)
 
 
+_SUPPRESSED_LOG_EVERY = 20
+
+
 class PollingPriceManager:
     """Polling-based price manager using Finnhub or yfinance."""
 
@@ -43,6 +46,7 @@ class PollingPriceManager:
         self.latest_prices: Dict[str, float] = {}
         self.open_prices: Dict[str, float] = {}
         self.price_callbacks: List[Callable] = []
+        self._failure_counts: Dict[str, int] = {}
 
         self.running = False
         self._thread: Optional[threading.Thread] = None
@@ -77,6 +81,8 @@ class PollingPriceManager:
         for symbol in self.subscribed_symbols:
             try:
                 quote_data = self._fetch_quote(symbol)
+                if not isinstance(quote_data, dict):
+                    raise ValueError(f"{symbol}: Empty quote payload")
 
                 current_price = quote_data.get("c")
                 open_price = quote_data.get("o")
@@ -103,6 +109,13 @@ class PollingPriceManager:
                 )
 
                 self.latest_prices[symbol] = current_price
+                previous_failures = self._failure_counts.pop(symbol, 0)
+                if previous_failures > 0:
+                    logger.info(
+                        "%s quote polling recovered after %d consecutive failures",
+                        symbol,
+                        previous_failures,
+                    )
 
                 price_data = {
                     "symbol": symbol,
@@ -128,7 +141,20 @@ class PollingPriceManager:
                 )
 
             except Exception as e:
-                logger.error(f"Failed to fetch {symbol} price: {e}")
+                failure_count = self._failure_counts.get(symbol, 0) + 1
+                self._failure_counts[symbol] = failure_count
+                message = f"Failed to fetch {symbol} price: {e}"
+
+                if failure_count == 1:
+                    logger.warning(message)
+                elif failure_count % _SUPPRESSED_LOG_EVERY == 0:
+                    logger.warning(
+                        "%s (repeated %d times; suppressing intermediate failures)",
+                        message,
+                        failure_count,
+                    )
+                else:
+                    logger.debug(message)
 
     def _fetch_quote(self, symbol: str) -> Dict[str, float]:
         """Fetch a normalized quote payload from the configured provider."""
@@ -136,7 +162,10 @@ class PollingPriceManager:
             return self._fetch_yfinance_quote(symbol)
         if not self.finnhub_client:
             raise ValueError("Finnhub API key required for finnhub polling")
-        return self.finnhub_client.quote(symbol)
+        quote = self.finnhub_client.quote(symbol)
+        if not isinstance(quote, dict):
+            raise ValueError(f"{symbol}: Invalid Finnhub quote payload")
+        return quote
 
     def _fetch_yfinance_quote(self, symbol: str) -> Dict[str, float]:
         """Fetch quote data from yfinance and normalize to Finnhub-like keys."""
@@ -162,6 +191,8 @@ class PollingPriceManager:
 
         if current_price is None:
             history = ticker.history(period="1d", interval="1m", auto_adjust=False)
+            if history is None:
+                raise ValueError(f"{symbol}: yfinance returned no history frame")
             if history.empty:
                 raise ValueError(f"{symbol}: No yfinance quote data")
             latest = history.iloc[-1]

backend/domains/news.py

@@ -30,6 +30,7 @@ def ensure_news_fresh(
     *,
     ticker: str,
     target_date: str | None = None,
+    refresh_if_stale: bool = True,
 ) -> dict[str, Any]:
     """Refresh raw news incrementally when stored watermarks are stale."""
     normalized_target = str(target_date or "").strip()[:10]
@@ -44,7 +45,7 @@ def ensure_news_fresh(
     watermarks = store.get_ticker_watermarks(ticker)
     last_news_fetch = str(watermarks.get("last_news_fetch") or "").strip()[:10]
     refreshed = False
-    if not last_news_fetch or last_news_fetch < normalized_target:
+    if refresh_if_stale and (not last_news_fetch or last_news_fetch < normalized_target):
         update_ticker_incremental(
             ticker,
             end_date=normalized_target,
@@ -69,8 +70,14 @@ def get_enriched_news(
     start_date: str | None = None,
     end_date: str | None = None,
     limit: int = 100,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=end_date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=end_date,
+        refresh_if_stale=refresh_if_stale,
+    )
     rows = store.get_news_items_enriched(
         ticker,
         start_date=start_date,
@@ -100,8 +107,14 @@ def get_news_for_date(
     ticker: str,
     date: str,
     limit: int = 20,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=date,
+        refresh_if_stale=refresh_if_stale,
+    )
     rows = store.get_news_items_enriched(
         ticker,
         trade_date=date,
@@ -129,8 +142,14 @@ def get_news_timeline(
     ticker: str,
     start_date: str,
     end_date: str,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=end_date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=end_date,
+        refresh_if_stale=refresh_if_stale,
+    )
     timeline = store.get_news_timeline_enriched(
         ticker,
         start_date=start_date,
@@ -165,8 +184,14 @@ def get_news_categories(
     start_date: str | None = None,
     end_date: str | None = None,
     limit: int = 200,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=end_date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=end_date,
+        refresh_if_stale=refresh_if_stale,
+    )
     rows = store.get_news_items_enriched(
         ticker,
         start_date=start_date,
@@ -196,8 +221,14 @@ def get_similar_days_payload(
     ticker: str,
     date: str,
     n_similar: int = 5,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=date,
+        refresh_if_stale=refresh_if_stale,
+    )
     result = find_similar_days(
         store,
         symbol=ticker,
@@ -213,8 +244,14 @@ def get_story_payload(
     *,
     ticker: str,
     as_of_date: str,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=as_of_date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=as_of_date,
+        refresh_if_stale=refresh_if_stale,
+    )
     enrich_news_for_symbol(
         store,
         ticker,
@@ -238,8 +275,14 @@ def get_range_explain_payload(
     end_date: str,
     article_ids: list[str] | None = None,
     limit: int = 100,
+    refresh_if_stale: bool = False,
 ) -> dict[str, Any]:
-    freshness = ensure_news_fresh(store, ticker=ticker, target_date=end_date)
+    freshness = ensure_news_fresh(
+        store,
+        ticker=ticker,
+        target_date=end_date,
+        refresh_if_stale=refresh_if_stale,
+    )
     news_rows = []
     if article_ids:
         news_rows = store.get_news_by_ids_enriched(ticker, article_ids)

backend/gateway_server.py

@@ -43,6 +43,72 @@ logger = logging.getLogger(__name__)
 _prompt_loader = get_prompt_loader()
 
 
+INFO_LOGGER_PREFIXES = (
+    "backend.agents",
+    "backend.core.pipeline",
+    "backend.core.scheduler",
+    "backend.services.gateway_cycle_support",
+)
+
+NOISY_LOGGER_LEVELS = {
+    "aiohttp": logging.WARNING,
+    "asyncio": logging.WARNING,
+    "dashscope": logging.WARNING,
+    "finnhub": logging.WARNING,
+    "httpcore": logging.WARNING,
+    "httpx": logging.WARNING,
+    "urllib3": logging.WARNING,
+    "websockets": logging.WARNING,
+    "yfinance": logging.WARNING,
+    "backend.data.polling_price_manager": logging.WARNING,
+    "backend.services.gateway": logging.WARNING,
+    "backend.services.market": logging.WARNING,
+    "backend.services.storage": logging.WARNING,
+}
+
+
+class SuppressNoisyInfoFilter(logging.Filter):
+    """Filter out low-signal library INFO logs while keeping warnings/errors."""
+
+    def filter(self, record: logging.LogRecord) -> bool:
+        message = record.getMessage()
+        if record.name == "httpx" and message.startswith("HTTP Request:"):
+            return False
+        if record.name.startswith("websockets") and "connection open" in message:
+            return False
+        if record.name.startswith("websockets") and "opening handshake failed" in message:
+            return False
+
+        if record.levelno >= logging.WARNING:
+            return True
+
+        return True
+
+
+def configure_gateway_logging(verbose: bool = False) -> None:
+    """Configure gateway logging with low-noise defaults for runtime logs."""
+    root_level = logging.DEBUG if verbose else logging.WARNING
+    logging.basicConfig(
+        level=root_level,
+        format="%(asctime)s | %(levelname)-7s | %(name)s:%(lineno)d - %(message)s",
+        force=True,
+    )
+
+    if not verbose:
+        suppress_filter = SuppressNoisyInfoFilter()
+        for handler in logging.getLogger().handlers:
+            handler.addFilter(suppress_filter)
+
+    for logger_name, level in NOISY_LOGGER_LEVELS.items():
+        logging.getLogger(logger_name).setLevel(logging.DEBUG if verbose else level)
+
+    if not verbose:
+        for prefix in INFO_LOGGER_PREFIXES:
+            logging.getLogger(prefix).setLevel(logging.INFO)
+
+    logging.getLogger(__name__).setLevel(logging.INFO if not verbose else logging.DEBUG)
+
+
 async def run_gateway(
     run_id: str,
     run_dir: Path,
@@ -52,7 +118,7 @@ async def run_gateway(
     """Run Gateway with Pipeline."""
 
     # Extract config
-    tickers = bootstrap.get("tickers", ["AAPL", "MSFT"])
+    tickers = bootstrap.get("tickers", ["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA", "AMD", "NFLX", "AVGO", "PLTR", "COIN"])
     initial_cash = float(bootstrap.get("initial_cash", 100000.0))
     margin_requirement = float(bootstrap.get("margin_requirement", 0.0))
     max_comm_cycles = int(bootstrap.get("max_comm_cycles", 2))
@@ -65,10 +131,8 @@ async def run_gateway(
     end_date = bootstrap.get("end_date")
     enable_memory = bootstrap.get("enable_memory", False)
     poll_interval = int(bootstrap.get("poll_interval", 10))
-    enable_mock = bootstrap.get("enable_mock", False)
 
     is_backtest = mode == "backtest"
-    is_mock = enable_mock or mode == "mock" or (not is_backtest and os.getenv("MOCK_MODE", "false").lower() == "true")
 
     logger.info(f"[Gateway Server] Starting run {run_id} on port {port}")
 
@@ -87,9 +151,8 @@ async def run_gateway(
             market_service = MarketService(
                 tickers=tickers,
                 poll_interval=poll_interval,
-                mock_mode=is_mock and not is_backtest,
                 backtest_mode=is_backtest,
-                api_key=os.getenv("FINNHUB_API_KEY") if not is_mock and not is_backtest else None,
+                api_key=os.getenv("FINNHUB_API_KEY") if not is_backtest else None,
                 backtest_start_date=start_date if is_backtest else None,
                 backtest_end_date=end_date if is_backtest else None,
             )
@@ -182,7 +245,6 @@ async def run_gateway(
                 scheduler_callback=scheduler_callback,
                 config={
                     "mode": mode,
-                    "mock_mode": is_mock,
                     "backtest_mode": is_backtest,
                     "tickers": tickers,
                     "config_name": run_id,
@@ -222,11 +284,7 @@ def main():
     args = parser.parse_args()
 
     # Setup logging
-    level = logging.DEBUG if args.verbose else logging.INFO
+    configure_gateway_logging(verbose=args.verbose)
-    logging.basicConfig(
-        level=level,
-        format="%(asctime)s | %(levelname)-7s | %(name)s:%(lineno)d - %(message)s",
-    )
 
     # Parse bootstrap
     bootstrap = json.loads(args.bootstrap)

backend/llm/models.py

@@ -3,6 +3,8 @@
 AgentScope Native Model Factory
 Uses native AgentScope model classes for LLM calls
 """
+import asyncio
+import inspect
 import os
 import time
 import logging
@@ -34,6 +36,27 @@ logger = logging.getLogger(__name__)
 T = TypeVar("T")
 
 
+def _usage_value(usage: Any, key: str, default: Any = 0) -> Any:
+    """Read usage fields from both object-style and dict-style usage payloads."""
+    if usage is None:
+        return default
+    if isinstance(usage, dict):
+        return usage.get(key, default)
+    try:
+        return getattr(usage, key)
+    except (AttributeError, KeyError):
+        return default
+
+
+def _usage_total_tokens(usage: Any) -> int:
+    total = _usage_value(usage, "total_tokens", None)
+    if total is not None:
+        return int(total or 0)
+    input_tokens = _usage_value(usage, "input_tokens", 0)
+    output_tokens = _usage_value(usage, "output_tokens", 0)
+    return int((input_tokens or 0) + (output_tokens or 0))
+
+
 class RetryChatModel:
     """Wraps an AgentScope model with automatic retry for transient errors.
 
@@ -55,6 +78,7 @@ class RetryChatModel:
         "502",
         "504",
         "connection",
+        "disconnected",
         "temporary",
         "overloaded",
         "too_many_requests",
@@ -150,8 +174,8 @@ class RetryChatModel:
                 # Track usage if available
                 if hasattr(result, "usage") and result.usage:
                     usage = result.usage
-                    self._total_tokens_used += getattr(usage, "total_tokens", 0)
+                    self._total_tokens_used += _usage_total_tokens(usage)
-                    self._total_cost += getattr(usage, "cost", 0.0)
+                    self._total_cost += float(_usage_value(usage, "cost", 0.0) or 0.0)
 
                 return result
 
@@ -192,9 +216,66 @@ class RetryChatModel:
             raise last_error
         raise RuntimeError("RetryChatModel: Unexpected state, no error but no result")
 
+    async def _call_with_retry_async(self, func: Callable[..., T], *args, **kwargs) -> T:
+        """Call an async function with retry logic for transient errors."""
+        last_error: Optional[Exception] = None
+
+        for attempt in range(1, self._max_retries + 1):
+            try:
+                result = await func(*args, **kwargs)
+
+                if hasattr(result, "usage") and result.usage:
+                    usage = result.usage
+                    self._total_tokens_used += _usage_total_tokens(usage)
+                    self._total_cost += float(_usage_value(usage, "cost", 0.0) or 0.0)
+
+                return result
+
+            except Exception as e:
+                last_error = e
+
+                if attempt >= self._max_retries:
+                    logger.error(
+                        "RetryChatModel: Max retries (%d) exhausted for %s",
+                        self._max_retries,
+                        self.model_name,
+                    )
+                    break
+
+                if not self._is_transient_error(e):
+                    logger.warning(
+                        "RetryChatModel: Non-transient error, not retrying: %s",
+                        str(e),
+                    )
+                    break
+
+                delay = self._calculate_delay(attempt)
+                logger.warning(
+                    "RetryChatModel: Transient async error on attempt %d/%d, "
+                    "retrying in %.1fs: %s",
+                    attempt,
+                    self._max_retries,
+                    delay,
+                    str(e)[:200],
+                )
+
+                if self._on_retry:
+                    self._on_retry(attempt, e, delay)
+
+                await asyncio.sleep(delay)
+
+        if last_error is not None:
+            raise last_error
+        raise RuntimeError("RetryChatModel: Unexpected async state, no error but no result")
+
     def __call__(self, *args, **kwargs) -> Any:
         """Forward calls to the wrapped model with retry logic."""
-        return self._call_with_retry(self._model, *args, **kwargs)
+        model_call = getattr(self._model, "__call__", None)
+        if inspect.iscoroutinefunction(self._model) or inspect.iscoroutinefunction(model_call):
+            return self._call_with_retry_async(self._model, *args, **kwargs)
+
+        result = self._model(*args, **kwargs)
+        return result
 
     def __getattr__(self, name: str) -> Any:
         """Proxy attribute access to the wrapped model."""
@@ -248,10 +329,18 @@ class TokenRecordingModelWrapper:
         if usage is None:
             return
 
-        self._prompt_tokens += getattr(usage, "prompt_tokens", 0)
+        prompt_tokens = _usage_value(usage, "prompt_tokens", None)
-        self._completion_tokens += getattr(usage, "completion_tokens", 0)
+        completion_tokens = _usage_value(usage, "completion_tokens", None)
-        self._total_tokens += getattr(usage, "total_tokens", 0)
+
-        self._total_cost += getattr(usage, "cost", 0.0)
+        if prompt_tokens is None:
+            prompt_tokens = _usage_value(usage, "input_tokens", 0)
+        if completion_tokens is None:
+            completion_tokens = _usage_value(usage, "output_tokens", 0)
+
+        self._prompt_tokens += int(prompt_tokens or 0)
+        self._completion_tokens += int(completion_tokens or 0)
+        self._total_tokens += _usage_total_tokens(usage)
+        self._total_cost += float(_usage_value(usage, "cost", 0.0) or 0.0)
 
     def __call__(self, *args, **kwargs) -> Any:
         """Forward calls and record usage."""
@@ -401,7 +490,8 @@ def create_model(
         if host:
             model_kwargs["host"] = host
 
-    return model_class(**model_kwargs)
+    model = model_class(**model_kwargs)
+    return RetryChatModel(model)
 
 
 def get_agent_model(agent_id: str, stream: bool = False):

backend/main.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 """
 Main Entry Point
-Supports: backtest, live, mock modes
+Supports: backtest, live modes
 """
 import argparse
 import asyncio
@@ -29,6 +29,7 @@ from backend.runtime.manager import (
     set_global_runtime_manager,
     clear_global_runtime_manager,
 )
+from backend.gateway_server import configure_gateway_logging
 from backend.services.gateway import Gateway
 from backend.services.market import MarketService
 from backend.services.storage import StorageService
@@ -38,6 +39,7 @@ load_dotenv()
 logger = logging.getLogger(__name__)
 loguru.logger.disable("flowllm")
 loguru.logger.disable("reme_ai")
+configure_gateway_logging(verbose=os.getenv("LOG_LEVEL", "").upper() == "DEBUG")
 _prompt_loader = get_prompt_loader()
 
 
@@ -226,17 +228,13 @@ async def run_with_gateway(args):
     )
     runtime_manager.prepare_run()
     set_global_runtime_manager(runtime_manager)
-    register_runtime_manager(runtime_manager)
 
     # Create market service
     market_service = MarketService(
         tickers=tickers,
         poll_interval=args.poll_interval,
-        mock_mode=args.mock and not is_backtest,
         backtest_mode=is_backtest,
-        api_key=os.getenv("FINNHUB_API_KEY")
+        api_key=os.getenv("FINNHUB_API_KEY") if not is_backtest else None,
-        if not args.mock and not is_backtest
-        else None,
         backtest_start_date=args.start_date if is_backtest else None,
         backtest_end_date=args.end_date if is_backtest else None,
     )
@@ -321,7 +319,6 @@ async def run_with_gateway(args):
         scheduler_callback=scheduler_callback,
         config={
             "mode": args.mode,
-            "mock_mode": args.mock,
             "backtest_mode": is_backtest,
             "tickers": tickers,
             "config_name": config_name,
@@ -354,8 +351,7 @@ def main():
     """Main entry point"""
     parser = argparse.ArgumentParser(description="Trading System")
     parser.add_argument("--mode", choices=["live", "backtest"], default="live")
-    parser.add_argument("--mock", action="store_true")
+    parser.add_argument("--config-name", default="live")
-    parser.add_argument("--config-name", default="mock")
     parser.add_argument("--host", default="0.0.0.0")
     parser.add_argument("--port", type=int, default=8765)
     parser.add_argument(

backend/runtime/manager.py

@@ -13,15 +13,30 @@ from .registry import RuntimeRegistry
 _global_runtime_manager: Optional["TradingRuntimeManager"] = None
 _shutdown_event: Optional[asyncio.Event] = None
 
+# Lazy import to avoid circular dependency
+_api_runtime = None
+
+
+def _get_api_runtime():
+    global _api_runtime
+    if _api_runtime is None:
+        from backend.api import runtime as api_runtime_module
+        _api_runtime = api_runtime_module
+    return _api_runtime
+
 
 def set_global_runtime_manager(manager: "TradingRuntimeManager") -> None:
     global _global_runtime_manager
     _global_runtime_manager = manager
+    # Sync to RuntimeState for consistency
+    _get_api_runtime().register_runtime_manager(manager)
 
 
 def clear_global_runtime_manager() -> None:
     global _global_runtime_manager
     _global_runtime_manager = None
+    # Sync to RuntimeState for consistency
+    _get_api_runtime().unregister_runtime_manager()
 
 
 def get_global_runtime_manager() -> Optional["TradingRuntimeManager"]:

backend/services/gateway.py

@@ -16,7 +16,6 @@ from websockets.asyncio.server import ServerConnection
 from backend.data.provider_utils import normalize_symbol
 from backend.domains import news as news_domain
 from backend.llm.models import get_agent_model_info
-from backend.utils.terminal_dashboard import get_dashboard
 from backend.core.pipeline import TradingPipeline
 from backend.core.state_sync import StateSync
 from backend.services.market import MarketService
@@ -26,10 +25,12 @@ from backend.tools.technical_signals import StockTechnicalAnalyzer
 from backend.core.scheduler import Scheduler
 from backend.services import gateway_admin_handlers
 from backend.services import gateway_cycle_support
+from backend.services import gateway_openclaw_handlers
 from backend.services import gateway_runtime_support
 from backend.services import gateway_stock_handlers
 from shared.client import NewsServiceClient
 from shared.client import TradingServiceClient
+from shared.client.openclaw_websocket_client import OpenClawWebSocketClient, DEFAULT_GATEWAY_URL as OPENCLAW_WS_URL
 
 logger = logging.getLogger(__name__)
 EDITABLE_AGENT_WORKSPACE_FILES = {
@@ -38,9 +39,6 @@ EDITABLE_AGENT_WORKSPACE_FILES = {
     "AGENTS.md",
     "MEMORY.md",
     "POLICY.md",
-    "HEARTBEAT.md",
-    "ROLE.md",
-    "STYLE.md",
 }
 
 
@@ -82,7 +80,6 @@ class Gateway:
         self._manual_cycle_task: Optional[asyncio.Task] = None
         self._backtest_start_date: Optional[str] = None
         self._backtest_end_date: Optional[str] = None
-        self._dashboard = get_dashboard()
         self._market_status_task: Optional[asyncio.Task] = None
         self._watchlist_ingest_task: Optional[asyncio.Task] = None
 
@@ -92,6 +89,7 @@ class Gateway:
         self._loop: Optional[asyncio.AbstractEventLoop] = None
         self._project_root = Path(__file__).resolve().parents[2]
         self._technical_analyzer = StockTechnicalAnalyzer()
+        self._openclaw_ws: OpenClawWebSocketClient | None = None
 
     async def start(self, host: str = "0.0.0.0", port: int = 8766):
         """Start gateway server with proper initialization order.
@@ -104,31 +102,11 @@ class Gateway:
         self._loop = asyncio.get_running_loop()
         self._provider_router.add_listener(self._on_provider_usage_changed)
 
-        # Initialize terminal dashboard
-        self._dashboard.set_config(
-            mode=self.mode,
-            config_name=self.config.get("config_name", "default"),
-            host=host,
-            port=port,
-            poll_interval=self.config.get("poll_interval", 10),
-            mock=self.config.get("mock_mode", False),
-            tickers=self.config.get("tickers", []),
-            initial_cash=self.storage.initial_cash,
-            start_date=self._backtest_start_date or "",
-            end_date=self._backtest_end_date or "",
-            data_sources=self._provider_router.get_usage_snapshot(),
-        )
-        self._dashboard.start()
-
         self.state_sync.load_state()
         self.market_service.set_price_recorder(self.storage.record_price_point)
         self.state_sync.update_state("status", "initializing")
         self.state_sync.update_state("server_mode", self.mode)
         self.state_sync.update_state("is_backtest", self.is_backtest)
-        self.state_sync.update_state(
-            "is_mock_mode",
-            self.config.get("mock_mode", False),
-        )
         self.state_sync.update_state("tickers", self.config.get("tickers", []))
         self.state_sync.update_state(
             "runtime_config",
@@ -152,18 +130,9 @@ class Gateway:
         )
 
         # Load and display existing portfolio state if available
-        summary = self.storage.load_file("summary")
+        dashboard_snapshot = self.storage.build_dashboard_snapshot_from_state(self.state_sync.state)
+        summary = dashboard_snapshot.get("summary")
         if summary:
-            holdings = self.storage.load_file("holdings") or []
-            trades = self.storage.load_file("trades") or []
-            current_date = self.state_sync.state.get("current_date")
-            self._dashboard.update(
-                date=current_date or "-",
-                status="running",
-                portfolio=summary,
-                holdings=holdings,
-                trades=trades,
-            )
             logger.info(
                 "Loaded existing portfolio: $%s",
                 f"{summary.get('totalAssetValue', 0):,.2f}",
@@ -189,6 +158,20 @@ class Gateway:
         # Give a brief moment for any existing clients to reconnect
         await asyncio.sleep(0.1)
 
+        # Connect to OpenClaw Gateway (18789) via WebSocket
+        logger.info("Connecting to OpenClaw Gateway...")
+        try:
+            self._openclaw_ws = OpenClawWebSocketClient(
+                url=OPENCLAW_WS_URL,
+                client_name="gateway-client",
+                client_version="1.0.0",
+            )
+            await self._openclaw_ws.connect()
+            logger.info("OpenClaw Gateway WebSocket connected")
+        except Exception as e:
+            logger.warning("Failed to connect to OpenClaw Gateway: %s", e)
+            self._openclaw_ws = None
+
         # ======================================================================
         # PHASE 2: Start market data service
         # Now frontend is connected, start pushing price updates
@@ -239,7 +222,6 @@ class Gateway:
     def _on_provider_usage_changed(self, snapshot: Dict[str, Any]):
         """Handle provider routing updates from the shared router."""
         self.state_sync.update_state("data_sources", snapshot)
-        self._dashboard.update(data_sources=snapshot)
         if self._loop and self._loop.is_running():
             asyncio.run_coroutine_threadsafe(
                 self.broadcast(
@@ -438,11 +420,77 @@ class Gateway:
                     await self._handle_get_stock_technical_indicators(websocket, data)
                 elif msg_type == "run_stock_enrich":
                     await self._handle_run_stock_enrich(websocket, data)
+                elif msg_type == "get_openclaw_status":
+                    await self._handle_get_openclaw_status(websocket, data)
+                elif msg_type == "get_openclaw_sessions":
+                    await self._handle_get_openclaw_sessions(websocket, data)
+                elif msg_type == "get_openclaw_session_detail":
+                    await self._handle_get_openclaw_session_detail(websocket, data)
+                elif msg_type == "get_openclaw_session_history":
+                    await self._handle_get_openclaw_session_history(websocket, data)
+                elif msg_type == "get_openclaw_cron":
+                    await self._handle_get_openclaw_cron(websocket, data)
+                elif msg_type == "get_openclaw_approvals":
+                    await self._handle_get_openclaw_approvals(websocket, data)
+                elif msg_type == "get_openclaw_agents":
+                    await self._handle_get_openclaw_agents(websocket, data)
+                elif msg_type == "get_openclaw_agents_presence":
+                    await self._handle_get_openclaw_agents_presence(websocket, data)
+                elif msg_type == "get_openclaw_skills":
+                    await self._handle_get_openclaw_skills(websocket, data)
+                elif msg_type == "get_openclaw_models":
+                    await self._handle_get_openclaw_models(websocket, data)
+                elif msg_type == "get_openclaw_hooks":
+                    await gateway_openclaw_handlers.handle_get_openclaw_hooks(self, websocket, data)
+                elif msg_type == "get_openclaw_plugins":
+                    await gateway_openclaw_handlers.handle_get_openclaw_plugins(self, websocket, data)
+                elif msg_type == "get_openclaw_secrets_audit":
+                    await gateway_openclaw_handlers.handle_get_openclaw_secrets_audit(self, websocket, data)
+                elif msg_type == "get_openclaw_security_audit":
+                    await gateway_openclaw_handlers.handle_get_openclaw_security_audit(self, websocket, data)
+                elif msg_type == "get_openclaw_daemon_status":
+                    await gateway_openclaw_handlers.handle_get_openclaw_daemon_status(self, websocket, data)
+                elif msg_type == "get_openclaw_pairing":
+                    await gateway_openclaw_handlers.handle_get_openclaw_pairing(self, websocket, data)
+                elif msg_type == "get_openclaw_qr":
+                    await gateway_openclaw_handlers.handle_get_openclaw_qr(self, websocket, data)
+                elif msg_type == "get_openclaw_update_status":
+                    await gateway_openclaw_handlers.handle_get_openclaw_update_status(self, websocket, data)
+                elif msg_type == "get_openclaw_models_aliases":
+                    await gateway_openclaw_handlers.handle_get_openclaw_models_aliases(self, websocket, data)
+                elif msg_type == "get_openclaw_models_fallbacks":
+                    await gateway_openclaw_handlers.handle_get_openclaw_models_fallbacks(self, websocket, data)
+                elif msg_type == "get_openclaw_models_image_fallbacks":
+                    await gateway_openclaw_handlers.handle_get_openclaw_models_image_fallbacks(self, websocket, data)
+                elif msg_type == "get_openclaw_skill_update":
+                    await gateway_openclaw_handlers.handle_get_openclaw_skill_update(self, websocket, data)
+                elif msg_type == "get_openclaw_workspace_files":
+                    await gateway_openclaw_handlers.handle_get_openclaw_workspace_files(self, websocket, data)
+                elif msg_type == "get_openclaw_workspace_file":
+                    await gateway_openclaw_handlers.handle_get_openclaw_workspace_file(self, websocket, data)
+                elif msg_type == "openclaw_resolve_session":
+                    await gateway_openclaw_handlers.handle_openclaw_resolve_session(self, websocket, data)
+                elif msg_type == "openclaw_create_session":
+                    await gateway_openclaw_handlers.handle_openclaw_create_session(self, websocket, data)
+                elif msg_type == "openclaw_send_message":
+                    await gateway_openclaw_handlers.handle_openclaw_send_message(self, websocket, data)
+                elif msg_type == "openclaw_subscribe_session":
+                    await gateway_openclaw_handlers.handle_openclaw_subscribe_session(self, websocket, data)
+                elif msg_type == "openclaw_unsubscribe_session":
+                    await gateway_openclaw_handlers.handle_openclaw_unsubscribe_session(self, websocket, data)
+                elif msg_type == "openclaw_reset_session":
+                    await gateway_openclaw_handlers.handle_openclaw_reset_session(self, websocket, data)
+                elif msg_type == "openclaw_delete_session":
+                    await gateway_openclaw_handlers.handle_openclaw_delete_session(self, websocket, data)
 
         except websockets.ConnectionClosed:
             pass
         except json.JSONDecodeError:
             pass
+        finally:
+            subscriber_map = getattr(self, "_openclaw_session_subscribers", None)
+            if isinstance(subscriber_map, dict):
+                subscriber_map.pop(websocket, None)
 
     async def _handle_get_stock_history(
         self,
@@ -544,13 +592,13 @@ class Gateway:
         websocket: ServerConnection,
         data: Dict[str, Any],
     ) -> None:
-        """Run one live/mock trading cycle on demand."""
+        """Run one live trading cycle on demand."""
         if self.is_backtest:
             await websocket.send(
                 json.dumps(
                     {
                         "type": "error",
-                        "message": "Manual trigger is only available in live/mock mode.",
+                        "message": "Manual trigger is only available in live mode.",
                     },
                     ensure_ascii=False,
                 ),
@@ -673,6 +721,83 @@ class Gateway:
     ) -> None:
         await gateway_admin_handlers.handle_update_agent_workspace_file(self, websocket, data)
 
+    async def _handle_get_openclaw_status(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_status(self, websocket, data)
+
+    async def _handle_get_openclaw_sessions(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_sessions(self, websocket, data)
+
+    async def _handle_get_openclaw_session_detail(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_session_detail(self, websocket, data)
+
+    async def _handle_get_openclaw_session_history(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_session_history(self, websocket, data)
+
+    async def _handle_get_openclaw_cron(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_cron(self, websocket, data)
+
+    async def _handle_get_openclaw_approvals(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_approvals(self, websocket, data)
+
+    async def _handle_get_openclaw_agents(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_agents(self, websocket, data)
+
+    async def _handle_get_openclaw_agents_presence(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_agents_presence(self, websocket, data)
+
+    async def _handle_get_openclaw_skills(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_skills(self, websocket, data)
+
+    async def _handle_get_openclaw_models(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_models(self, websocket, data)
+
+    async def _handle_get_openclaw_workspace_files(
+        self,
+        websocket: ServerConnection,
+        data: Dict[str, Any],
+    ) -> None:
+        await gateway_openclaw_handlers.handle_get_openclaw_workspace_files(self, websocket, data)
+
     @staticmethod
     def _normalize_watchlist(raw_tickers: Any) -> List[str]:
         return gateway_runtime_support.normalize_watchlist(raw_tickers)

backend/services/gateway_admin_handlers.py

@@ -1,5 +1,12 @@
 # -*- coding: utf-8 -*-
-"""Runtime/workspace/skills handlers extracted from the main Gateway module."""
+"""Runtime/workspace/skills handlers extracted from the main Gateway module.
+
+Deprecated note:
+    Agent/workspace/skill read-write operations are being migrated to
+    agent_service REST endpoints. These websocket handlers remain as a
+    compatibility fallback and should not be considered the primary control
+    plane path for frontend reads/writes.
+"""
 
 from __future__ import annotations
 

backend/services/gateway_cycle_support.py

@@ -7,7 +7,7 @@ import asyncio
 import logging
 from typing import Any
 
-from backend.data.market_ingest import ingest_symbols
+from backend.data.market_ingest import ingest_symbols, refresh_news_for_symbols
 from backend.domains import trading as trading_domain
 from backend.utils.msg_adapter import FrontendAdapter
 
@@ -61,7 +61,7 @@ async def market_status_monitor(gateway: Any) -> None:
             status = gateway.market_service.get_market_status()
             if status["status"] == "open" and not gateway.storage.is_live_session_active:
                 gateway.storage.start_live_session()
-                summary = gateway.storage.load_file("summary") or {}
+                summary = gateway.storage.build_dashboard_snapshot_from_state(gateway.state_sync.state).get("summary") or {}
                 gateway._session_start_portfolio_value = summary.get(
                     "totalAssetValue",
                     gateway.storage.initial_cash,
@@ -147,25 +147,10 @@ async def on_heartbeat_trigger(gateway: Any, date: str) -> None:
 
     for analyst in analysts:
         try:
-            ws_id = getattr(analyst, "workspace_id", None)
+            logger.debug(
-            if ws_id:
+                "[Heartbeat] No heartbeat configured for %s, skipping",
-                from backend.agents.workspace_manager import get_workspace_dir
+                analyst.name,
-                from pathlib import Path
+            )
-                from agentscope.message import Msg
-
-                ws_dir = get_workspace_dir(ws_id)
-                if ws_dir:
-                    hb_path = Path(ws_dir) / "HEARTBEAT.md"
-                    if hb_path.exists():
-                        content = hb_path.read_text(encoding="utf-8").strip()
-                        if content:
-                            hb_task = f"# 定期主动检查\n\n{content}\n\n请执行上述检查并报告结果。"
-                            logger.info("[Heartbeat] Running heartbeat for %s", analyst.name)
-                            msg = Msg(role="user", content=hb_task, name="system")
-                            await analyst.reply([msg])
-                            logger.info("[Heartbeat] %s heartbeat complete", analyst.name)
-                            continue
-            logger.debug("[Heartbeat] No HEARTBEAT.md for %s, skipping", analyst.name)
         except Exception as exc:
             logger.error("[Heartbeat] %s failed: %s", analyst.name, exc, exc_info=True)
 
@@ -175,7 +160,6 @@ async def run_backtest_cycle(gateway: Any, date: str, tickers: list[str]) -> Non
     await gateway.market_service.emit_market_open()
 
     await gateway.state_sync.on_cycle_start(date)
-    gateway._dashboard.update(date=date, status="Analyzing...")
 
     prices = gateway.market_service.get_open_prices()
     close_prices = gateway.market_service.get_close_prices()
@@ -200,8 +184,24 @@ async def run_live_cycle(gateway: Any, date: str, tickers: list[str]) -> None:
     trading_date = gateway.market_service.get_live_trading_date()
     logger.info("Live cycle: triggered=%s, trading_date=%s", date, trading_date)
 
+    try:
+        news_refresh = await asyncio.to_thread(
+            refresh_news_for_symbols,
+            tickers,
+            end_date=trading_date,
+            store=gateway.storage.market_store,
+        )
+        logger.info(
+            "News refresh complete: %s",
+            ", ".join(
+                f"{item['symbol']} news={item['news']}"
+                for item in news_refresh
+            ) or "no symbols",
+        )
+    except Exception as exc:
+        logger.warning("Live cycle news refresh failed: %s", exc)
+
     await gateway.state_sync.on_cycle_start(trading_date)
-    gateway._dashboard.update(date=trading_date, status="Analyzing...")
 
     market_caps = await get_market_caps(gateway, tickers, trading_date)
     schedule_mode = gateway.config.get("schedule_mode", "daily")
@@ -240,17 +240,15 @@ async def run_live_cycle(gateway: Any, date: str, tickers: list[str]) -> None:
 
 
 async def finalize_cycle(gateway: Any, date: str) -> None:
-    summary = gateway.storage.load_file("summary") or {}
+    dashboard_snapshot = gateway.storage.build_dashboard_snapshot_from_state(gateway.state_sync.state)
+    summary = dashboard_snapshot.get("summary") or {}
     if gateway.storage.is_live_session_active:
         summary.update(gateway.storage.get_live_returns())
 
     await gateway.state_sync.on_cycle_end(date, portfolio_summary=summary)
-    holdings = gateway.storage.load_file("holdings") or []
+    leaderboard = dashboard_snapshot.get("leaderboard") or []
-    trades = gateway.storage.load_file("trades") or []
-    leaderboard = gateway.storage.load_file("leaderboard") or []
     if leaderboard:
         await gateway.state_sync.on_leaderboard_update(leaderboard)
-    gateway._dashboard.update(date=date, status="Running", portfolio=summary, holdings=holdings, trades=trades)
 
 
 async def get_market_caps(gateway: Any, tickers: list[str], date: str) -> dict[str, float]:
@@ -311,24 +309,16 @@ def save_cycle_results(
 
 async def run_backtest_dates(gateway: Any, dates: list[str]) -> None:
     gateway.state_sync.set_backtest_dates(dates)
-    gateway._dashboard.update(days_total=len(dates), days_completed=0)
     await gateway.state_sync.on_system_message(f"Starting backtest - {len(dates)} trading days")
     try:
-        for i, date in enumerate(dates):
+        for date in dates:
-            gateway._dashboard.update(days_completed=i)
             await gateway.on_strategy_trigger(date=date)
             await asyncio.sleep(0.1)
         await gateway.state_sync.on_system_message(f"Backtest complete - {len(dates)} days")
-        summary = gateway.storage.load_file("summary") or {}
-        gateway._dashboard.update(status="Complete", portfolio=summary, days_completed=len(dates))
-        gateway._dashboard.stop()
-        gateway._dashboard.print_final_summary()
     except Exception as exc:
         error_msg = f"Backtest failed: {type(exc).__name__}: {str(exc)}"
         logger.error(error_msg, exc_info=True)
         asyncio.create_task(gateway.state_sync.on_system_message(error_msg))
-        gateway._dashboard.update(status=f"Failed: {str(exc)}")
-        gateway._dashboard.stop()
         raise
     finally:
         gateway._backtest_task = None
@@ -358,7 +348,6 @@ def set_backtest_dates(gateway: Any, dates: list[str]) -> None:
     if dates:
         gateway._backtest_start_date = dates[0]
         gateway._backtest_end_date = dates[-1]
-        gateway._dashboard.days_total = len(dates)
 
 
 def stop_gateway(gateway: Any) -> None:
@@ -370,4 +359,14 @@ def stop_gateway(gateway: Any) -> None:
         gateway._market_status_task.cancel()
     if gateway._watchlist_ingest_task:
         gateway._watchlist_ingest_task.cancel()
-    gateway._dashboard.stop()
+    # Close OpenClaw WebSocket connection
+    if gateway._openclaw_ws:
+        import asyncio
+        try:
+            loop = asyncio.get_event_loop()
+            if loop.is_running():
+                loop.create_task(gateway._openclaw_ws.disconnect())
+            else:
+                loop.run_until_complete(gateway._openclaw_ws.disconnect())
+        except Exception:
+            pass

backend/services/gateway_openclaw_handlers.py

@@ -0,0 +1,534 @@
+# -*- coding: utf-8 -*-
+"""OpenClaw WebSocket handlers — gateway calls OpenClaw Gateway via WebSocket."""
+
+from __future__ import annotations
+
+import json
+import logging
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from backend.services.gateway import Gateway
+
+logger = logging.getLogger(__name__)
+
+
+def _ensure_session_bridge(gateway) -> None:
+    """Forward OpenClaw session events into 大时代 frontend websockets."""
+    if getattr(gateway, "_openclaw_session_bridge_ready", False):
+        return
+
+    async def _forward(event) -> None:
+        payload = event.payload or {}
+        session_key = str(payload.get("sessionKey") or payload.get("key") or "").strip()
+        if not session_key:
+            return
+
+        subscriber_map = getattr(gateway, "_openclaw_session_subscribers", {})
+        targets = [
+            ws
+            for ws, session_keys in list(subscriber_map.items())
+            if session_key in session_keys
+        ]
+        if not targets:
+            return
+
+        message = json.dumps(
+            {
+                "type": "openclaw_session_event",
+                "event": event.event,
+                "session_key": session_key,
+                "payload": payload,
+            }
+        )
+        stale = []
+        for ws in targets:
+            try:
+                await ws.send(message)
+            except Exception:
+                stale.append(ws)
+
+        for ws in stale:
+            try:
+                subscriber_map.pop(ws, None)
+            except Exception:
+                pass
+
+    def _handler(event) -> None:
+        try:
+            import asyncio
+
+            asyncio.create_task(_forward(event))
+        except Exception as exc:
+            logger.debug("OpenClaw session bridge skipped event: %s", exc)
+
+    client = _get_ws_client(gateway)
+    client.add_event_handler(_handler)
+    gateway._openclaw_session_bridge_ready = True
+    gateway._openclaw_session_bridge_handler = _handler
+    if not hasattr(gateway, "_openclaw_session_subscribers"):
+        gateway._openclaw_session_subscribers = {}
+
+
+def _get_ws_client(gateway) -> "OpenClawWebSocketClient":
+    """Get the OpenClaw WebSocket client from gateway."""
+    from shared.client.openclaw_websocket_client import OpenClawWebSocketClient
+    client = gateway._openclaw_ws
+    if client is None:
+        raise RuntimeError("OpenClaw Gateway not connected")
+    return client
+
+
+async def _ws_call(gateway, method: str, params: dict | None = None) -> dict:
+    """Call OpenClaw Gateway via WebSocket and return result."""
+    try:
+        client = _get_ws_client(gateway)
+        return await client.call_method(method, params)
+    except Exception as exc:
+        logger.warning("OpenClaw Gateway call failed for %s: %s", method, exc)
+        return {"error": str(exc)[:200]}
+
+
+async def handle_get_openclaw_status(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "status")
+    await websocket.send(json.dumps({"type": "openclaw_status_loaded", "data": result}))
+
+
+async def handle_get_openclaw_sessions(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "sessions.list", {"limit": 50, "includeLastMessage": True})
+    await websocket.send(json.dumps({"type": "openclaw_sessions_loaded", "data": result}))
+
+
+async def handle_get_openclaw_session_detail(gateway, websocket, data: dict) -> None:
+    session_key = data.get("session_key", "")
+    result = await _ws_call(gateway, "sessions.list", {"limit": 200, "includeLastMessage": True})
+    session = None
+    if isinstance(result, dict):
+        for item in result.get("sessions", []) or []:
+            if not isinstance(item, dict):
+                continue
+            if item.get("key") == session_key or item.get("sessionKey") == session_key:
+                session = item
+                break
+    await websocket.send(json.dumps({
+        "type": "openclaw_session_detail_loaded",
+        "data": {"session": session, "error": None if session else f"session '{session_key}' not found"},
+        "session_key": session_key,
+    }))
+
+
+async def handle_get_openclaw_session_history(gateway, websocket, data: dict) -> None:
+    session_key = data.get("session_key", "")
+    limit = data.get("limit", 20)
+    try:
+        from backend.services.openclaw_cli import OpenClawCliService
+
+        result = OpenClawCliService().get_session_history_model(session_key, limit=limit)
+        payload = {
+            "session_key": result.session_key,
+            "session_id": result.session_id,
+            "history": result.events,
+            "events": result.events,
+            "raw_text": result.raw_text,
+        }
+    except Exception as exc:
+        payload = {"error": str(exc)[:200], "history": []}
+    await websocket.send(json.dumps({
+        "type": "openclaw_session_history_loaded",
+        "data": payload,
+        "session_key": session_key,
+    }))
+
+
+async def handle_openclaw_resolve_session(gateway, websocket, data: dict) -> None:
+    params = {}
+    agent_id = str(data.get("agent_id") or "").strip()
+    label = str(data.get("label") or "").strip()
+    channel = str(data.get("channel") or "").strip()
+    if agent_id:
+        params["agentId"] = agent_id
+    if label:
+        params["label"] = label
+    if channel:
+        params["channel"] = channel
+    params["includeGlobal"] = bool(data.get("include_global", True))
+    result = await _ws_call(gateway, "sessions.resolve", params)
+    await websocket.send(json.dumps({"type": "openclaw_session_resolved", "data": result}))
+
+
+async def handle_openclaw_create_session(gateway, websocket, data: dict) -> None:
+    params = {}
+    agent_id = str(data.get("agent_id") or "").strip()
+    label = str(data.get("label") or "").strip()
+    model = str(data.get("model") or "").strip()
+    initial_message = str(data.get("initial_message") or "").strip()
+    if agent_id:
+        params["agentId"] = agent_id
+    if label:
+        params["label"] = label
+    if model:
+        params["model"] = model
+    if initial_message:
+        params["message"] = initial_message
+    result = await _ws_call(gateway, "sessions.create", params)
+    await websocket.send(json.dumps({"type": "openclaw_session_created", "data": result}))
+
+
+async def handle_openclaw_send_message(gateway, websocket, data: dict) -> None:
+    session_key = str(data.get("session_key") or "").strip()
+    message = str(data.get("message") or "").strip()
+    thinking = str(data.get("thinking") or "").strip()
+    if not session_key or not message:
+        await websocket.send(
+            json.dumps(
+                {
+                    "type": "openclaw_message_sent",
+                    "data": {"error": "session_key and message are required"},
+                }
+            )
+        )
+        return
+
+    params = {"key": session_key, "message": message}
+    if thinking:
+        params["thinking"] = thinking
+    result = await _ws_call(gateway, "sessions.send", params)
+    await websocket.send(
+        json.dumps(
+            {
+                "type": "openclaw_message_sent",
+                "data": result,
+                "session_key": session_key,
+            }
+        )
+    )
+
+
+async def handle_openclaw_subscribe_session(gateway, websocket, data: dict) -> None:
+    session_key = str(data.get("session_key") or "").strip()
+    if not session_key:
+        await websocket.send(
+            json.dumps(
+                {
+                    "type": "openclaw_session_subscribed",
+                    "data": {"error": "session_key is required"},
+                }
+            )
+        )
+        return
+
+    _ensure_session_bridge(gateway)
+    result = await _ws_call(gateway, "sessions.messages.subscribe", {"key": session_key})
+    if not isinstance(result, dict) or not result.get("error"):
+        subscriber_map = getattr(gateway, "_openclaw_session_subscribers", {})
+        subscriber_map.setdefault(websocket, set()).add(session_key)
+        gateway._openclaw_session_subscribers = subscriber_map
+    await websocket.send(
+        json.dumps(
+            {
+                "type": "openclaw_session_subscribed",
+                "data": result,
+                "session_key": session_key,
+            }
+        )
+    )
+
+
+async def handle_openclaw_unsubscribe_session(gateway, websocket, data: dict) -> None:
+    session_key = str(data.get("session_key") or "").strip()
+    if not session_key:
+        await websocket.send(
+            json.dumps(
+                {
+                    "type": "openclaw_session_unsubscribed",
+                    "data": {"error": "session_key is required"},
+                }
+            )
+        )
+        return
+
+    result = await _ws_call(gateway, "sessions.messages.unsubscribe", {"key": session_key})
+    subscriber_map = getattr(gateway, "_openclaw_session_subscribers", {})
+    session_keys = subscriber_map.get(websocket)
+    if isinstance(session_keys, set):
+        session_keys.discard(session_key)
+        if not session_keys:
+            subscriber_map.pop(websocket, None)
+    gateway._openclaw_session_subscribers = subscriber_map
+    await websocket.send(
+        json.dumps(
+            {
+                "type": "openclaw_session_unsubscribed",
+                "data": result,
+                "session_key": session_key,
+            }
+        )
+    )
+
+
+async def handle_openclaw_reset_session(gateway, websocket, data: dict) -> None:
+    session_key = str(data.get("session_key") or "").strip()
+    if not session_key:
+        await websocket.send(
+            json.dumps(
+                {
+                    "type": "openclaw_session_reset",
+                    "data": {"error": "session_key is required"},
+                }
+            )
+        )
+        return
+
+    result = await _ws_call(gateway, "sessions.reset", {"key": session_key})
+    await websocket.send(
+        json.dumps(
+            {
+                "type": "openclaw_session_reset",
+                "data": result,
+                "session_key": session_key,
+            }
+        )
+    )
+
+
+async def handle_openclaw_delete_session(gateway, websocket, data: dict) -> None:
+    session_key = str(data.get("session_key") or "").strip()
+    if not session_key:
+        await websocket.send(
+            json.dumps(
+                {
+                    "type": "openclaw_session_deleted",
+                    "data": {"error": "session_key is required"},
+                }
+            )
+        )
+        return
+
+    result = await _ws_call(gateway, "sessions.delete", {"key": session_key})
+    await websocket.send(
+        json.dumps(
+            {
+                "type": "openclaw_session_deleted",
+                "data": result,
+                "session_key": session_key,
+            }
+        )
+    )
+
+
+async def handle_get_openclaw_cron(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "cron.list")
+    await websocket.send(json.dumps({"type": "openclaw_cron_loaded", "data": result}))
+
+
+async def handle_get_openclaw_approvals(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "exec.approvals.get")
+    await websocket.send(json.dumps({"type": "openclaw_approvals_loaded", "data": result}))
+
+
+async def handle_get_openclaw_agents(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "agents.list")
+    sessions_result = await _ws_call(
+        gateway,
+        "sessions.list",
+        {"limit": 200, "includeLastMessage": True},
+    )
+    config_result = await _ws_call(gateway, "config.get")
+    session_model_by_agent: dict[str, str] = {}
+    default_session_model: str | None = None
+    agent_skills_by_id: dict[str, list[str] | None] = {}
+    default_agent_skills: list[str] | None = None
+
+    parsed_config = config_result.get("parsed") if isinstance(config_result, dict) else None
+    if isinstance(parsed_config, dict):
+        agents_cfg = parsed_config.get("agents")
+        if isinstance(agents_cfg, dict):
+            defaults_cfg = agents_cfg.get("defaults")
+            if isinstance(defaults_cfg, dict):
+                default_skills = defaults_cfg.get("skills")
+                if isinstance(default_skills, list):
+                    default_agent_skills = [
+                        str(skill).strip()
+                        for skill in default_skills
+                        if str(skill).strip()
+                    ]
+            list_cfg = agents_cfg.get("list")
+            if isinstance(list_cfg, list):
+                for entry in list_cfg:
+                    if not isinstance(entry, dict):
+                        continue
+                    agent_id = str(entry.get("id") or "").strip()
+                    if not agent_id:
+                        continue
+                    skills = entry.get("skills")
+                    if isinstance(skills, list):
+                        agent_skills_by_id[agent_id] = [
+                            str(skill).strip()
+                            for skill in skills
+                            if str(skill).strip()
+                        ]
+                    elif skills == []:
+                        agent_skills_by_id[agent_id] = []
+
+    if isinstance(sessions_result, dict) and isinstance(sessions_result.get("sessions"), list):
+        defaults = sessions_result.get("defaults")
+        if isinstance(defaults, dict):
+            value = (
+                defaults.get("model")
+                or defaults.get("modelName")
+                or defaults.get("model_name")
+            )
+            if value:
+                default_session_model = str(value)
+        for session in sessions_result.get("sessions", []):
+            if not isinstance(session, dict):
+                continue
+            agent_id = str(
+                session.get("agentId")
+                or session.get("agent_id")
+                or ""
+            ).strip()
+            if not agent_id:
+                key = str(session.get("key") or session.get("sessionKey") or "").strip()
+                parts = key.split(":")
+                if len(parts) >= 3 and parts[0] == "agent":
+                    agent_id = parts[1]
+            model_value = (
+                session.get("model")
+                or session.get("modelName")
+                or session.get("model_name")
+                or session.get("resolvedModel")
+                or session.get("resolved_model")
+                or session.get("defaultModel")
+                or session.get("default_model")
+            )
+            if agent_id and model_value and agent_id not in session_model_by_agent:
+                session_model_by_agent[agent_id] = str(model_value)
+
+    if isinstance(result, dict) and isinstance(result.get("agents"), list):
+        normalized_agents = []
+        for agent in result.get("agents", []):
+            if not isinstance(agent, dict):
+                normalized_agents.append(agent)
+                continue
+            normalized = dict(agent)
+            if not normalized.get("model"):
+                normalized["model"] = (
+                    normalized.get("modelName")
+                    or normalized.get("model_name")
+                    or normalized.get("resolvedModel")
+                    or normalized.get("resolved_model")
+                    or normalized.get("defaultModel")
+                    or normalized.get("default_model")
+                    or session_model_by_agent.get(str(normalized.get("id") or "").strip())
+                    or default_session_model
+                )
+            agent_id = str(normalized.get("id") or "").strip()
+            if "skills" not in normalized:
+                normalized["skills"] = agent_skills_by_id.get(agent_id, default_agent_skills)
+            normalized_agents.append(normalized)
+        result = {**result, "agents": normalized_agents}
+    await websocket.send(json.dumps({"type": "openclaw_agents_loaded", "data": result}))
+
+
+async def handle_get_openclaw_agents_presence(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "node.list")
+    await websocket.send(json.dumps({"type": "openclaw_agents_presence_loaded", "data": result}))
+
+
+async def handle_get_openclaw_skills(gateway, websocket, data: dict) -> None:
+    agent_id = str(data.get("agent_id") or "").strip()
+    params = {"agentId": agent_id} if agent_id else {}
+    result = await _ws_call(gateway, "skills.status", params)
+    await websocket.send(json.dumps({"type": "openclaw_skills_loaded", "data": result}))
+
+
+async def handle_get_openclaw_models(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "models.list")
+    await websocket.send(json.dumps({"type": "openclaw_models_loaded", "data": result}))
+
+
+async def handle_get_openclaw_hooks(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "tools.catalog")
+    await websocket.send(json.dumps({"type": "openclaw_hooks_loaded", "data": result}))
+
+
+async def handle_get_openclaw_plugins(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "config.get")
+    await websocket.send(json.dumps({"type": "openclaw_plugins_loaded", "data": result}))
+
+
+async def handle_get_openclaw_secrets_audit(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "secrets.reload")
+    await websocket.send(json.dumps({"type": "openclaw_secrets_audit_loaded", "data": result}))
+
+
+async def handle_get_openclaw_security_audit(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "gateway.identity.get")
+    await websocket.send(json.dumps({"type": "openclaw_security_audit_loaded", "data": result}))
+
+
+async def handle_get_openclaw_daemon_status(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "doctor.memory.status")
+    await websocket.send(json.dumps({"type": "openclaw_daemon_status_loaded", "data": result}))
+
+
+async def handle_get_openclaw_pairing(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "device.pair.list")
+    await websocket.send(json.dumps({"type": "openclaw_pairing_loaded", "data": result}))
+
+
+async def handle_get_openclaw_qr(gateway, websocket, data: dict) -> None:
+    await websocket.send(json.dumps({"type": "openclaw_qr_loaded", "data": {"error": "QR code not available via WebSocket"}}))
+
+
+async def handle_get_openclaw_update_status(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "update.run")
+    await websocket.send(json.dumps({"type": "openclaw_update_status_loaded", "data": result}))
+
+
+async def handle_get_openclaw_models_aliases(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "models.list")
+    await websocket.send(json.dumps({"type": "openclaw_models_aliases_loaded", "data": result}))
+
+
+async def handle_get_openclaw_models_fallbacks(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "models.list")
+    await websocket.send(json.dumps({"type": "openclaw_models_fallbacks_loaded", "data": result}))
+
+
+async def handle_get_openclaw_models_image_fallbacks(gateway, websocket, data: dict) -> None:
+    result = await _ws_call(gateway, "models.list")
+    await websocket.send(json.dumps({"type": "openclaw_models_image_fallbacks_loaded", "data": result}))
+
+
+async def handle_get_openclaw_skill_update(gateway, websocket, data: dict) -> None:
+    slug = data.get("slug")
+    all_flag = data.get("all", False)
+    params = {}
+    if slug is not None:
+        params["slug"] = slug
+    if all_flag:
+        params["all"] = "true"
+    result = await _ws_call(gateway, "skills.update", params)
+    await websocket.send(json.dumps({"type": "openclaw_skill_update_loaded", "data": result}))
+
+
+async def handle_get_openclaw_workspace_files(gateway, websocket, data: dict) -> None:
+    raw_workspace = data.get("workspace", "")
+    # Use the workspace param (which is actually the agent.id from frontend) as agent_id
+    agent_id = raw_workspace or "main"
+    result = await _ws_call(gateway, "agents.files.list", {"agentId": agent_id})
+    if isinstance(result, dict):
+        result["workspace"] = agent_id
+    await websocket.send(json.dumps({"type": "openclaw_workspace_files_loaded", "data": result}))
+
+
+async def handle_get_openclaw_workspace_file(gateway, websocket, data: dict) -> None:
+    agent_id = data.get("agent_id", "main")
+    file_name = data.get("file_name", "")
+    if not file_name:
+        await websocket.send(json.dumps({"type": "openclaw_workspace_file_loaded", "data": {"error": "file_name is required"}}))
+        return
+    result = await _ws_call(gateway, "agents.files.get", {"agentId": agent_id, "name": file_name})
+    await websocket.send(json.dumps({"type": "openclaw_workspace_file_loaded", "data": result}))

backend/services/gateway_runtime_support.py

@@ -141,7 +141,7 @@ def apply_runtime_config(gateway: Any, runtime_config: dict[str, Any]) -> dict[s
 
 
 def sync_runtime_state(gateway: Any) -> None:
-    """Refresh persisted state and dashboard after runtime config changes."""
+    """Refresh persisted state after runtime config changes."""
     gateway.state_sync.update_state("tickers", gateway.config.get("tickers", []))
     gateway.state_sync.update_state(
         "runtime_config",
@@ -159,16 +159,3 @@ def sync_runtime_state(gateway: Any) -> None:
 
     gateway.storage.update_server_state_from_dashboard(gateway.state_sync.state)
     gateway.state_sync.save_state()
-
-    gateway._dashboard.tickers = list(gateway.config.get("tickers", []))
-    gateway._dashboard.initial_cash = gateway.storage.initial_cash
-    gateway._dashboard.enable_memory = bool(gateway.config.get("enable_memory", False))
-
-    summary = gateway.storage.load_file("summary") or {}
-    holdings = gateway.storage.load_file("holdings") or []
-    trades = gateway.storage.load_file("trades") or []
-    gateway._dashboard.update(
-        portfolio=summary,
-        holdings=holdings,
-        trades=trades,
-    )

backend/services/gateway_stock_handlers.py

@@ -152,6 +152,7 @@ async def handle_get_stock_news(gateway: Any, websocket: Any, data: dict[str, An
             start_date=start_date,
             end_date=end_date,
             limit=max(limit, 50),
+            refresh_if_stale=False,
         )
         news_rows = (payload.get("news") or [])[-limit:]
         source = "market_store"
@@ -202,6 +203,7 @@ async def handle_get_stock_news_for_date(gateway: Any, websocket: Any, data: dic
             ticker=ticker,
             date=trade_date,
             limit=limit,
+            refresh_if_stale=False,
         )
         news_rows = payload.get("news") or []
         source = "market_store"
@@ -255,6 +257,7 @@ async def handle_get_stock_news_timeline(gateway: Any, websocket: Any, data: dic
             ticker=ticker,
             start_date=start_date,
             end_date=end_date,
+            refresh_if_stale=False,
         )
         timeline = payload.get("timeline") or []
 
@@ -313,6 +316,7 @@ async def handle_get_stock_news_categories(gateway: Any, websocket: Any, data: d
             start_date=start_date,
             end_date=end_date,
             limit=200,
+            refresh_if_stale=False,
         )
         categories = payload.get("categories") or {}
 
@@ -361,6 +365,7 @@ async def handle_get_stock_range_explain(gateway: Any, websocket: Any, data: dic
             end_date=end_date,
             article_ids=article_ids if isinstance(article_ids, list) else None,
             limit=100,
+            refresh_if_stale=False,
         )
         result = payload.get("result")
 

backend/services/market.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 """
 Market Data Service
-Supports live, mock, and backtest modes
+Supports live and backtest modes
 """
 import asyncio
 import logging
@@ -10,7 +10,7 @@ from typing import Any, Callable, Dict, List, Optional
 from zoneinfo import ZoneInfo
 
 import pandas_market_calendars as mcal
-from backend.config.data_config import get_data_source
+from backend.config.data_config import get_data_sources
 from backend.data.provider_utils import normalize_symbol
 
 logger = logging.getLogger(__name__)
@@ -36,7 +36,6 @@ class MarketService:
         self,
         tickers: List[str],
         poll_interval: int = 10,
-        mock_mode: bool = False,
         backtest_mode: bool = False,
         api_key: Optional[str] = None,
         backtest_start_date: Optional[str] = None,
@@ -44,7 +43,6 @@ class MarketService:
     ):
         self.tickers = [normalize_symbol(ticker) for ticker in tickers]
         self.poll_interval = poll_interval
-        self.mock_mode = mock_mode
         self.backtest_mode = backtest_mode
         self.api_key = api_key
         self.backtest_start_date = backtest_start_date
@@ -69,8 +67,6 @@ class MarketService:
         """Return the active live quote provider for UI/debugging."""
         if self.backtest_mode:
             return "backtest"
-        if self.mock_mode:
-            return "mock"
         if self._price_manager and hasattr(self._price_manager, "provider"):
             provider = getattr(self._price_manager, "provider", None)
             if isinstance(provider, str) and provider.strip():
@@ -81,8 +77,6 @@ class MarketService:
     def mode_name(self) -> str:
         if self.backtest_mode:
             return "BACKTEST"
-        elif self.mock_mode:
-            return "MOCK"
         return "LIVE"
 
     async def start(self, broadcast_func: Callable):
@@ -96,8 +90,6 @@ class MarketService:
 
         if self.backtest_mode:
             self._start_backtest_mode()
-        elif self.mock_mode:
-            self._start_mock_mode()
         else:
             self._start_real_mode()
 
@@ -125,26 +117,10 @@ class MarketService:
 
         return callback
 
-    def _start_mock_mode(self):
-        from backend.data.mock_price_manager import MockPriceManager
-
-        self._price_manager = MockPriceManager(
-            poll_interval=self.poll_interval,
-            volatility=0.5,
-        )
-        self._price_manager.add_price_callback(self._make_price_callback())
-        self._price_manager.subscribe(
-            self.tickers,
-            base_prices={t: 100.0 for t in self.tickers},
-        )
-        self._price_manager.start()
-
     def _start_real_mode(self):
         from backend.data.polling_price_manager import PollingPriceManager
 
-        provider = get_data_source()
+        provider = self._resolve_live_quote_provider()
-        if provider == "local_csv":
-            provider = "yfinance"
 
         if provider == "finnhub" and not self.api_key:
             raise ValueError("API key required for live mode")
@@ -157,6 +133,13 @@ class MarketService:
         self._price_manager.subscribe(self.tickers)
         self._price_manager.start()
 
+    def _resolve_live_quote_provider(self) -> str:
+        """Pick the first configured provider that supports live quote polling."""
+        for provider in get_data_sources():
+            if provider in {"finnhub", "yfinance"}:
+                return provider
+        return "yfinance"
+
     def _start_backtest_mode(self):
         from backend.data.historical_price_manager import (
             HistoricalPriceManager,
@@ -257,12 +240,6 @@ class MarketService:
             if removed:
                 self._price_manager.unsubscribe(removed)
             if added:
-                if self.mock_mode:
-                    self._price_manager.subscribe(
-                        added,
-                        base_prices={ticker: 100.0 for ticker in added},
-                    )
-                else:
                 self._price_manager.subscribe(added)
 
             if self.backtest_mode and self._current_date:

backend/services/openclaw_cli.py

@@ -0,0 +1,754 @@
+# -*- coding: utf-8 -*-
+"""Thin service wrapper around the OpenClaw CLI."""
+
+from __future__ import annotations
+
+import json
+import os
+import shlex
+import shutil
+import subprocess
+import sys
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any
+
+from shared.models.openclaw import (
+    AgentSummary,
+    AgentsList,
+    ApprovalRequest,
+    ApprovalsList,
+    CronJob,
+    CronList,
+    DaemonStatus,
+    HookStatusEntry,
+    HookStatusReport,
+    ModelAliasesList,
+    ModelFallbacksList,
+    ModelRow,
+    ModelsList,
+    OpenClawStatus,
+    PairingListResponse,
+    PluginDiagnostic,
+    PluginRecord,
+    PluginsList,
+    QrCodeResponse,
+    SecretsAuditReport,
+    SecurityAuditResponse,
+    SecurityAuditReport,
+    SessionEntry,
+    SessionHistory,
+    SessionsList,
+    SkillStatusEntry,
+    SkillStatusReport,
+    SkillUpdateResult,
+    UpdateCheckResult,
+    UpdateStatusResponse,
+    normalize_agents,
+    normalize_approvals,
+    normalize_cron_jobs,
+    normalize_daemon_status,
+    normalize_hooks,
+    normalize_model_aliases,
+    normalize_model_fallbacks,
+    normalize_models,
+    normalize_pairing,
+    normalize_plugins,
+    normalize_qr,
+    normalize_security_audit,
+    normalize_secrets_audit,
+    normalize_session_history,
+    normalize_sessions,
+    normalize_skill_update,
+    normalize_skills,
+    normalize_status,
+    normalize_update_status,
+)
+
+
+PROJECT_ROOT = Path(__file__).resolve().parents[2]
+REFERENCE_OPENCLAW_ROOT = PROJECT_ROOT / "reference" / "openclaw"
+REFERENCE_OPENCLAW_ENTRY = REFERENCE_OPENCLAW_ROOT / "openclaw.mjs"
+
+
+class OpenClawCliError(RuntimeError):
+    """Raised when the OpenClaw CLI invocation fails."""
+
+    def __init__(
+        self,
+        message: str,
+        *,
+        command: list[str],
+        exit_code: int | None = None,
+        stdout: str = "",
+        stderr: str = "",
+    ) -> None:
+        super().__init__(message)
+        self.command = command
+        self.exit_code = exit_code
+        self.stdout = stdout
+        self.stderr = stderr
+
+
+@dataclass(frozen=True)
+class OpenClawCliResult:
+    """Command execution result."""
+
+    command: list[str]
+    exit_code: int
+    stdout: str
+    stderr: str
+
+
+def resolve_openclaw_base_command() -> list[str]:
+    """Resolve the command prefix used to launch OpenClaw."""
+    explicit = os.getenv("OPENCLAW_CMD", "").strip()
+    if explicit:
+        return shlex.split(explicit)
+
+    installed = shutil.which("openclaw")
+    if installed:
+        return [installed]
+
+    if REFERENCE_OPENCLAW_ENTRY.exists():
+        return [sys.executable if sys.executable.endswith("node") else "node", str(REFERENCE_OPENCLAW_ENTRY)]
+
+    return ["openclaw"]
+
+
+def resolve_openclaw_cwd() -> Path:
+    """Resolve the working directory for CLI execution."""
+    explicit = os.getenv("OPENCLAW_CWD", "").strip()
+    if explicit:
+        return Path(explicit).expanduser()
+    if REFERENCE_OPENCLAW_ROOT.exists():
+        return REFERENCE_OPENCLAW_ROOT
+    return PROJECT_ROOT
+
+
+class OpenClawCliService:
+    """OpenClaw CLI integration service."""
+
+    def __init__(
+        self,
+        *,
+        base_command: list[str] | None = None,
+        cwd: Path | None = None,
+        timeout_seconds: float | None = None,
+    ) -> None:
+        self.base_command = list(base_command or resolve_openclaw_base_command())
+        self.cwd = cwd or resolve_openclaw_cwd()
+        self.timeout_seconds = timeout_seconds or float(
+            os.getenv("OPENCLAW_TIMEOUT_SECONDS", "15")
+        )
+
+    def health(self) -> dict[str, Any]:
+        """Return the current CLI wiring state."""
+        binary = self.base_command[0] if self.base_command else "openclaw"
+        resolved = shutil.which(binary) if len(self.base_command) == 1 else binary
+        return {
+            "status": "healthy",
+            "service": "openclaw-service",
+            "base_command": self.base_command,
+            "cwd": str(self.cwd),
+            "binary_resolved": resolved is not None,
+            "reference_entry_available": REFERENCE_OPENCLAW_ENTRY.exists(),
+            "timeout_seconds": self.timeout_seconds,
+        }
+
+    def status(self) -> dict[str, Any]:
+        """Read `openclaw status --json`."""
+        return self.run_json(["status", "--json"])
+
+    def list_sessions(self) -> dict[str, Any]:
+        """Read `openclaw sessions --json`."""
+        return self.run_json(["sessions", "--json"])
+
+    def get_session(self, session_key: str) -> dict[str, Any]:
+        """Resolve a single session out of the sessions list."""
+        payload = self.list_sessions()
+        sessions = payload.get("sessions") or []
+        for item in sessions:
+            if not isinstance(item, dict):
+                continue
+            if item.get("key") == session_key or item.get("sessionKey") == session_key:
+                return item
+        raise KeyError(session_key)
+
+    def get_session_history(self, session_key: str, *, limit: int = 20) -> dict[str, Any]:
+        """Read session history with a JSON-first fallback to raw text."""
+        args = ["sessions", "history", session_key, "--json", "--limit", str(limit)]
+        try:
+            return self.run_json(args)
+        except OpenClawCliError as exc:
+            raise exc
+        except json.JSONDecodeError:
+            result = self.run(args)
+            return {
+                "sessionKey": session_key,
+                "limit": limit,
+                "rawText": result.stdout,
+            }
+
+    def list_cron_jobs(self) -> dict[str, Any]:
+        """Read `openclaw cron list --json`."""
+        return self.run_json(["cron", "list", "--json"])
+
+    def list_approvals(self) -> dict[str, Any]:
+        """Read `openclaw approvals get --json`."""
+        return self.run_json(["approvals", "get", "--json"])
+
+    def list_agents(self) -> dict[str, Any]:
+        """Read `openclaw agents list --json`."""
+        return self.run_json(["agents", "list", "--json"])
+
+    def list_skills(self) -> dict[str, Any]:
+        """Read `openclaw skills list --json`."""
+        return self.run_json(["skills", "list", "--json"])
+
+    def list_models(self) -> dict[str, Any]:
+        """Read `openclaw models list --json`."""
+        return self.run_json(["models", "list", "--json"])
+
+    def list_hooks(self) -> dict[str, Any]:
+        """Read `openclaw hooks list --json`."""
+        return self.run_json(["hooks", "list", "--json"])
+
+    def list_plugins(self) -> dict[str, Any]:
+        """Read `openclaw plugins list --json`."""
+        return self.run_json(["plugins", "list", "--json"])
+
+    def secrets_audit(self) -> dict[str, Any]:
+        """Read `openclaw secrets audit --json`."""
+        return self.run_json(["secrets", "audit", "--json"])
+
+    def security_audit(self) -> dict[str, Any]:
+        """Read `openclaw security audit --json`."""
+        return self.run_json(["security", "audit", "--json"])
+
+    def daemon_status(self) -> dict[str, Any]:
+        """Read `openclaw daemon status --json`."""
+        return self.run_json(["daemon", "status", "--json"])
+
+    def pairing_list(self) -> dict[str, Any]:
+        """Read `openclaw pairing list --json`."""
+        return self.run_json(["pairing", "list", "--json"])
+
+    def qr_code(self) -> dict[str, Any]:
+        """Read `openclaw qr --json`."""
+        return self.run_json(["qr", "--json"])
+
+    def update_status(self) -> dict[str, Any]:
+        """Read `openclaw update status --json`."""
+        return self.run_json(["update", "status", "--json"])
+
+    def list_model_aliases(self) -> dict[str, Any]:
+        """Read `openclaw models aliases list --json`."""
+        return self.run_json(["models", "aliases", "list", "--json"])
+
+    def list_model_fallbacks(self) -> dict[str, Any]:
+        """Read `openclaw models fallbacks list --json`."""
+        return self.run_json(["models", "fallbacks", "list", "--json"])
+
+    def list_model_image_fallbacks(self) -> dict[str, Any]:
+        """Read `openclaw models image-fallbacks list --json`."""
+        return self.run_json(["models", "image-fallbacks", "list", "--json"])
+
+    def skill_update(self, *, slug: str | None = None, all: bool = False) -> dict[str, Any]:
+        """Read `openclaw skills update --json`."""
+        args = ["skills", "update", "--json"]
+        if slug:
+            args.append(slug)
+        if all:
+            args.append("--all")
+        return self.run_json(args)
+
+    def models_status(self, *, probe: bool = False) -> dict[str, Any]:
+        """Read `openclaw models status --json [--probe]`."""
+        args = ["models", "status", "--json"]
+        if probe:
+            args.append("--probe")
+        return self.run_json(args)
+
+    def channels_status(self, *, probe: bool = False) -> dict[str, Any]:
+        """Read `openclaw channels status [--probe] --json`."""
+        args = ["channels", "status", "--json"]
+        if probe:
+            args.append("--probe")
+        return self.run_json(args)
+
+    def list_workspace_files(self, workspace_path: str) -> dict[str, Any]:
+        """List .md files in an OpenClaw agent workspace with their content.
+
+        Reads the workspace directory and returns metadata + content for each .md file.
+        """
+        import json
+        from pathlib import Path
+
+        wp = Path(workspace_path).expanduser().resolve()
+        if not wp.exists() or not wp.is_dir():
+            return {"workspace": str(wp), "files": [], "error": "workspace not found"}
+
+        md_files = sorted(wp.glob("*.md"))
+        files = []
+        for md_file in md_files:
+            try:
+                content = md_file.read_text(encoding="utf-8")
+                # Preview: first 300 chars
+                preview = content[:300].strip()
+                files.append({
+                    "name": md_file.name,
+                    "path": str(md_file),
+                    "size": len(content),
+                    "preview": preview,
+                    "previewTruncated": len(content) > 300,
+                })
+            except OSError as exc:
+                files.append({
+                    "name": md_file.name,
+                    "path": str(md_file),
+                    "size": 0,
+                    "preview": "",
+                    "error": str(exc),
+                })
+
+        return {"workspace": str(wp), "files": files}
+
+    def channels_list(self) -> dict[str, Any]:
+        """Read `openclaw channels list --json`."""
+        return self.run_json(["channels", "list", "--json"])
+
+    def hook_info(self, name: str) -> dict[str, Any]:
+        """Read `openclaw hooks info <name> --json`."""
+        args = ["hooks", "info", name, "--json"]
+        try:
+            return self.run_json(args)
+        except json.JSONDecodeError:
+            result = self.run(args)
+            return {"raw": result.stdout}
+
+    def hooks_check(self) -> dict[str, Any]:
+        """Read `openclaw hooks check --json`."""
+        return self.run_json(["hooks", "check", "--json"])
+
+    def plugins_inspect(self, *, plugin_id: str | None = None, all: bool = False) -> dict[str, Any]:
+        """Read `openclaw plugins inspect [--json] [--all]`."""
+        args = ["plugins", "inspect", "--json"]
+        if all:
+            args.append("--all")
+        elif plugin_id:
+            args.append(plugin_id)
+        return self.run_json(args)
+
+    # -------------------------------------------------------------------------
+    # Typed variants — these use Pydantic models and are the preferred path.
+    # -------------------------------------------------------------------------
+
+    def status_model(self) -> OpenClawStatus:
+        """Read and parse `openclaw status --json` into a typed model."""
+        raw = self.status()
+        return normalize_status(raw)
+
+    def list_sessions_model(self) -> SessionsList:
+        """Read and parse `openclaw sessions --json` into a typed model."""
+        raw = self.list_sessions()
+        return normalize_sessions(raw)
+
+    def get_session_model(self, session_key: str) -> SessionEntry:
+        """Resolve a single session and return a typed model."""
+        raw = self.get_session(session_key)
+        return SessionEntry.model_validate(raw, strict=False)
+
+    def get_session_history_model(self, session_key: str, *, limit: int = 20) -> SessionHistory:
+        """Read session history and return a typed model."""
+        raw = self.get_session_history(session_key, limit=limit)
+        return normalize_session_history(raw, session_key=session_key)
+
+    def list_cron_jobs_model(self) -> CronList:
+        """Read and parse `openclaw cron list --json` into a typed model."""
+        raw = self.list_cron_jobs()
+        return normalize_cron_jobs(raw)
+
+    def list_approvals_model(self) -> ApprovalsList:
+        """Read and parse `openclaw approvals get --json` into a typed model."""
+        raw = self.list_approvals()
+        return normalize_approvals(raw)
+
+    # -------------------------------------------------------------------------
+    # Typed variants
+    # -------------------------------------------------------------------------
+
+    def list_agents_model(self) -> AgentsList:
+        """Read and parse `openclaw agents list --json` into a typed model."""
+        raw = self.list_agents()
+        if isinstance(raw, list):
+            return AgentsList(agents=[AgentSummary.model_validate(a, strict=False) for a in raw if isinstance(a, dict)])
+        return normalize_agents(raw)
+
+    def list_skills_model(self) -> SkillStatusReport:
+        """Read and parse `openclaw skills list --json` into a typed model."""
+        raw = self.list_skills()
+        return normalize_skills(raw)
+
+    def list_models_model(self) -> ModelsList:
+        """Read and parse `openclaw models list --json` into a typed model."""
+        raw = self.list_models()
+        if isinstance(raw, list):
+            return ModelsList(models=[ModelRow.model_validate(m, strict=False) for m in raw if isinstance(m, dict)])
+        return normalize_models(raw)
+
+    def list_hooks_model(self) -> HookStatusReport:
+        raw = self.list_hooks()
+        return normalize_hooks(raw)
+
+    def list_plugins_model(self) -> PluginsList:
+        raw = self.list_plugins()
+        return normalize_plugins(raw)
+
+    def secrets_audit_model(self) -> SecretsAuditReport:
+        raw = self.secrets_audit()
+        return normalize_secrets_audit(raw)
+
+    def security_audit_model(self) -> SecurityAuditResponse:
+        raw = self.security_audit()
+        return normalize_security_audit(raw)
+
+    def daemon_status_model(self) -> DaemonStatus:
+        raw = self.daemon_status()
+        return normalize_daemon_status(raw)
+
+    def pairing_list_model(self) -> PairingListResponse:
+        raw = self.pairing_list()
+        return normalize_pairing(raw)
+
+    def qr_code_model(self) -> QrCodeResponse:
+        raw = self.qr_code()
+        return normalize_qr(raw)
+
+    def update_status_model(self) -> UpdateStatusResponse:
+        raw = self.update_status()
+        return normalize_update_status(raw)
+
+    def list_model_aliases_model(self) -> ModelAliasesList:
+        raw = self.list_model_aliases()
+        return normalize_model_aliases(raw)
+
+    def list_model_fallbacks_model(self) -> ModelFallbacksList:
+        raw = self.list_model_fallbacks()
+        return normalize_model_fallbacks(raw)
+
+    def list_model_image_fallbacks_model(self) -> ModelFallbacksList:
+        raw = self.list_model_image_fallbacks()
+        return normalize_model_fallbacks(raw)
+
+    def skill_update_model(self, *, slug: str | None = None, all: bool = False) -> SkillUpdateResult:
+        raw = self.skill_update(slug=slug, all=all)
+        return normalize_skill_update(raw)
+
+    def models_status_model(self, *, probe: bool = False) -> dict[str, Any]:
+        """Read `openclaw models status --json` and return the raw dict."""
+        return self.models_status(probe=probe)
+
+    def channels_status_model(self, *, probe: bool = False) -> dict[str, Any]:
+        """Read `openclaw channels status --json` and return the raw dict."""
+        return self.channels_status(probe=probe)
+
+    def channels_list_model(self) -> dict[str, Any]:
+        """Read `openclaw channels list --json` and return the raw dict."""
+        return self.channels_list()
+
+    def hook_info_model(self, name: str) -> dict[str, Any]:
+        """Read `openclaw hooks info <name> --json` and return the raw dict."""
+        return self.hook_info(name)
+
+    def hooks_check_model(self) -> dict[str, Any]:
+        """Read `openclaw hooks check --json` and return the raw dict."""
+        return self.hooks_check()
+
+    def plugins_inspect_model(self, *, plugin_id: str | None = None, all: bool = False) -> dict[str, Any]:
+        """Read `openclaw plugins inspect --json [--all]` and return the raw dict."""
+        return self.plugins_inspect(plugin_id=plugin_id, all=all)
+
+    def agents_bindings(self, *, agent: str | None = None) -> dict[str, Any]:
+        """Read `openclaw agents bindings --json [--agent <id>]`."""
+        args = ["agents", "bindings", "--json"]
+        if agent:
+            args.extend(["--agent", agent])
+        return self.run_json(args)
+
+    def agents_bindings_model(self, *, agent: str | None = None) -> dict[str, Any]:
+        """Read `openclaw agents bindings --json` and return the raw dict."""
+        return self.agents_bindings(agent=agent)
+
+    def agents_presence(self) -> dict[str, Any]:
+        """Read session presence for all agents from runtime session files.
+
+        Reads ~/.openclaw/agents/{agentId}/sessions/sessions.json for each agent
+        and counts sessions in active states within a recency window.
+        """
+        import json
+        from pathlib import Path
+
+        openclaw_home = Path.home() / ".openclaw"
+        agents_path = openclaw_home / "agents"
+
+        if not agents_path.exists():
+            return {"status": "not_connected", "agents": {}}
+
+        ACTIVE_STATES = {
+            "running", "active", "busy", "blocked", "waiting_approval",
+            "working", "in_progress", "processing", "thinking", "executing", "streaming",
+        }
+
+        RECENCY_WINDOW_MS = 45 * 60 * 1000  # 45 minutes
+
+        result: dict[str, Any] = {"status": "connected", "agents": {}}
+
+        try:
+            for agent_dir in agents_path.iterdir():
+                if not agent_dir.is_dir():
+                    continue
+                sessions_file = agent_dir / "sessions" / "sessions.json"
+                if not sessions_file.exists():
+                    continue
+
+                try:
+                    sessions_data = json.loads(sessions_file.read_text())
+                except (json.JSONDecodeError, OSError):
+                    continue
+
+                sessions = sessions_data if isinstance(sessions_data, list) else []
+                now_ms = 0  # placeholder; we'll skip recency check if no ts field
+
+                active_count = 0
+                for session in sessions:
+                    if not isinstance(session, dict):
+                        continue
+                    state = str(session.get("state") or session.get("status") or "").lower()
+                    if state in ACTIVE_STATES:
+                        active_count += 1
+
+                if active_count > 0:
+                    result["agents"][agent_dir.name] = {
+                        "activeSessions": active_count,
+                        "status": "active",
+                    }
+                else:
+                    result["agents"][agent_dir.name] = {
+                        "activeSessions": 0,
+                        "status": "idle",
+                    }
+        except OSError:
+            result["status"] = "partial"
+
+        return result
+
+    def agents_from_config(self) -> dict[str, Any]:
+        """Read agent list directly from openclaw.json config file.
+
+        Falls back to scanning ~/.openclaw/agents/ directories when config is absent.
+        This avoids the CLI timeout from `agents list --json`.
+        """
+        import json
+
+        openclaw_home = Path.home() / ".openclaw"
+        config_path = openclaw_home / "openclaw.json"
+
+        if not config_path.exists():
+            return {"status": "not_connected", "agents": []}
+
+        try:
+            raw = json.loads(config_path.read_text())
+        except (json.JSONDecodeError, OSError):
+            return {"status": "partial", "agents": []}
+
+        agents_list = raw.get("agents", {}).get("list", [])
+        if not agents_list:
+            return {"status": "partial", "agents": [], "detail": "agents.list is empty"}
+
+        agents = []
+        for entry in agents_list:
+            if not isinstance(entry, dict):
+                continue
+            agent_id = entry.get("id", "").strip()
+            if not agent_id:
+                continue
+            agents.append({
+                "id": agent_id,
+                "name": entry.get("name", "").strip() or agent_id,
+                "model": entry.get("model") or "",
+                "workspace": entry.get("workspace") or "",
+                "is_default": entry.get("id") == raw.get("agents", {}).get("defaults", {}).get("id"),
+            })
+
+        return {"status": "connected", "agents": agents}
+
+    def gateway_status(self, *, url: str | None = None, token: str | None = None) -> dict[str, Any]:
+        """Read `openclaw gateway status --json [--url <url>] [--token <token>]`. May fail if gateway is unreachable."""
+        args = ["gateway", "status", "--json"]
+        if url:
+            args.extend(["--url", url])
+        if token:
+            args.extend(["--token", token])
+        return self.run_json(args)
+
+    def memory_status(self, *, agent: str | None = None, deep: bool = False) -> dict[str, Any]:
+        """Read `openclaw memory status --json [--agent <id>] [--deep]`. Returns array of per-agent status."""
+        args = ["memory", "status", "--json"]
+        if agent:
+            args.extend(["--agent", agent])
+        if deep:
+            args.append("--deep")
+        return self.run_json(args)
+
+    # -------------------------------------------------------------------------
+    # Write agents commands
+    # -------------------------------------------------------------------------
+
+    def agents_add(
+        self,
+        name: str,
+        *,
+        workspace: str | None = None,
+        model: str | None = None,
+        agent_dir: str | None = None,
+        bind: list[str] | None = None,
+        non_interactive: bool = False,
+    ) -> dict[str, Any]:
+        """Run `openclaw agents add <name> [--workspace <dir>] [--model <id>] [--agent-dir <dir>] [--bind <spec>] [--non-interactive] --json`."""
+        args = ["agents", "add", name, "--json"]
+        if workspace:
+            args.extend(["--workspace", workspace])
+        if model:
+            args.extend(["--model", model])
+        if agent_dir:
+            args.extend(["--agent-dir", agent_dir])
+        if bind:
+            for b in bind:
+                args.extend(["--bind", b])
+        if non_interactive:
+            args.append("--non-interactive")
+        return self.run_json(args)
+
+    def agents_delete(self, id: str, *, force: bool = False) -> dict[str, Any]:
+        """Run `openclaw agents delete <id> [--force] --json`."""
+        args = ["agents", "delete", id, "--json"]
+        if force:
+            args.append("--force")
+        return self.run_json(args)
+
+    def agents_bind(
+        self,
+        *,
+        agent: str | None = None,
+        bind: list[str] | None = None,
+    ) -> dict[str, Any]:
+        """Run `openclaw agents bind [--agent <id>] [--bind <spec>] --json`."""
+        args = ["agents", "bind", "--json"]
+        if agent:
+            args.extend(["--agent", agent])
+        if bind:
+            for b in bind:
+                args.extend(["--bind", b])
+        return self.run_json(args)
+
+    def agents_unbind(
+        self,
+        *,
+        agent: str | None = None,
+        bind: list[str] | None = None,
+        all: bool = False,
+    ) -> dict[str, Any]:
+        """Run `openclaw agents unbind [--agent <id>] [--bind <spec>] [--all] --json`."""
+        args = ["agents", "unbind", "--json"]
+        if agent:
+            args.extend(["--agent", agent])
+        if bind:
+            for b in bind:
+                args.extend(["--bind", b])
+        if all:
+            args.append("--all")
+        return self.run_json(args)
+
+    def agents_set_identity(
+        self,
+        *,
+        agent: str | None = None,
+        workspace: str | None = None,
+        identity_file: str | None = None,
+        name: str | None = None,
+        emoji: str | None = None,
+        theme: str | None = None,
+        avatar: str | None = None,
+        from_identity: bool = False,
+    ) -> dict[str, Any]:
+        """Run `openclaw agents set-identity [--agent <id>] [--workspace <dir>] [--identity-file <path>] [--from-identity] [--name <n>] [--emoji <e>] [--theme <t>] [--avatar <a>] --json`."""
+        args = ["agents", "set-identity", "--json"]
+        if agent:
+            args.extend(["--agent", agent])
+        if workspace:
+            args.extend(["--workspace", workspace])
+        if identity_file:
+            args.extend(["--identity-file", identity_file])
+        if from_identity:
+            args.append("--from-identity")
+        if name:
+            args.extend(["--name", name])
+        if emoji:
+            args.extend(["--emoji", emoji])
+        if theme:
+            args.extend(["--theme", theme])
+        if avatar:
+            args.extend(["--avatar", avatar])
+        return self.run_json(args)
+
+    def run_json(self, args: list[str]) -> dict[str, Any]:
+        """Run the CLI and decode JSON stdout, falling back to stderr."""
+        result = self.run(args)
+        text = result.stdout.strip() or result.stderr.strip()
+        if not text:
+            return {}
+        return json.loads(text)
+
+    def run(self, args: list[str]) -> OpenClawCliResult:
+        """Run the CLI and return stdout/stderr."""
+        command = [*self.base_command, *args]
+        env = os.environ.copy()
+        try:
+            completed = subprocess.run(
+                command,
+                cwd=self.cwd,
+                env=env,
+                capture_output=True,
+                text=True,
+                timeout=self.timeout_seconds,
+                check=False,
+            )
+        except FileNotFoundError as exc:
+            raise OpenClawCliError(
+                "OpenClaw CLI executable was not found.",
+                command=command,
+            ) from exc
+        except subprocess.TimeoutExpired as exc:
+            raise OpenClawCliError(
+                f"OpenClaw CLI timed out after {self.timeout_seconds:.1f}s.",
+                command=command,
+                stdout=exc.stdout or "",
+                stderr=exc.stderr or "",
+            ) from exc
+
+        if completed.returncode != 0:
+            raise OpenClawCliError(
+                "OpenClaw CLI command failed.",
+                command=command,
+                exit_code=completed.returncode,
+                stdout=completed.stdout,
+                stderr=completed.stderr,
+            )
+
+        return OpenClawCliResult(
+            command=command,
+            exit_code=completed.returncode,
+            stdout=completed.stdout,
+            stderr=completed.stderr,
+        )

backend/services/storage.py

@@ -11,7 +11,6 @@ from pathlib import Path
 from typing import Any, Dict, List, Optional
 
 from backend.data.market_store import MarketStore
-from .research_db import ResearchDb
 from .runtime_db import RuntimeDb
 
 logger = logging.getLogger(__name__)
@@ -22,19 +21,25 @@ class StorageService:
     Storage service for data persistence
 
     Responsibilities:
-    1. Load/save dashboard JSON files
+    1. Export dashboard JSON files
         (summary, holdings, stats, trades, leaderboard)
     2. Load/save internal state (_internal_state.json)
     3. Load/save server state (server_state.json) with feed history
     4. Manage portfolio state persistence
     5. Support loading from saved state to resume execution
+
+    Notes:
+    - team_dashboard/*.json is treated as an export/compatibility layer
+      rather than the authoritative runtime source of truth.
+    - authoritative runtime reads should prefer in-memory state, server_state,
+      runtime.db, and market_research.db.
     """
 
     def __init__(
         self,
         dashboard_dir: Path,
         initial_cash: float = 100000.0,
-        config_name: str = "mock",
+        config_name: str = "live",
     ):
         """
         Initialize storage service
@@ -49,7 +54,7 @@ class StorageService:
         self.initial_cash = initial_cash
         self.config_name = config_name
 
-        # Dashboard file paths
+        # Dashboard export file paths
         self.files = {
             "summary": self.dashboard_dir / "summary.json",
             "holdings": self.dashboard_dir / "holdings.json",
@@ -66,7 +71,6 @@ class StorageService:
         self.state_dir.mkdir(parents=True, exist_ok=True)
         self.server_state_file = self.state_dir / "server_state.json"
         self.runtime_db = RuntimeDb(self.state_dir / "runtime.db")
-        self.research_db = ResearchDb(self.state_dir / "research.db")
         self.market_store = MarketStore()
 
         # Feed history (for agent messages)
@@ -84,16 +88,8 @@ class StorageService:
 
         logger.info(f"Storage service initialized: {self.dashboard_dir}")
 
-    def load_file(self, file_type: str) -> Optional[Any]:
+    def load_export_file(self, file_type: str) -> Optional[Any]:
-        """
+        """Load dashboard export JSON file."""
-        Load dashboard JSON file
-
-        Args:
-            file_type: One of: summary, holdings, stats, trades, leaderboard
-
-        Returns:
-            Loaded data or None if file doesn't exist
-        """
         file_path = self.files.get(file_type)
         if not file_path or not file_path.exists():
             return None
@@ -105,14 +101,12 @@ class StorageService:
             logger.error(f"Failed to load {file_type}.json: {e}")
             return None
 
-    def save_file(self, file_type: str, data: Any):
+    def load_file(self, file_type: str) -> Optional[Any]:
-        """
+        """Backward-compatible alias for export-layer JSON reads."""
-        Save dashboard JSON file
+        return self.load_export_file(file_type)
 
-        Args:
+    def save_export_file(self, file_type: str, data: Any):
-            file_type: One of: summary, holdings, stats, trades, leaderboard
+        """Save dashboard export JSON file."""
-            data: Data to save
-        """
         file_path = self.files.get(file_type)
         if not file_path:
             logger.error(f"Unknown file type: {file_type}")
@@ -129,6 +123,48 @@ class StorageService:
         except Exception as e:
             logger.error(f"Failed to save {file_type}.json: {e}")
 
+    def save_file(self, file_type: str, data: Any):
+        """Backward-compatible alias for export-layer JSON writes."""
+        self.save_export_file(file_type, data)
+
+    def build_dashboard_snapshot_from_state(
+        self,
+        state: Optional[Dict[str, Any]] = None,
+    ) -> Dict[str, Any]:
+        """Build dashboard view data from runtime state instead of JSON exports."""
+        runtime_state = state or self.load_server_state()
+        portfolio = dict(runtime_state.get("portfolio") or {})
+        holdings = list(runtime_state.get("holdings") or [])
+        stats = runtime_state.get("stats") or self._get_default_stats()
+        trades = list(runtime_state.get("trades") or [])
+        leaderboard = list(runtime_state.get("leaderboard") or [])
+
+        summary = {
+            "totalAssetValue": portfolio.get("total_value", self.initial_cash),
+            "totalReturn": portfolio.get("pnl_percent", 0.0),
+            "cashPosition": portfolio.get("cash", self.initial_cash),
+            "tickerWeights": stats.get("tickerWeights", {}),
+            "totalTrades": len(trades),
+            "pnlPct": portfolio.get("pnl_percent", 0.0),
+            "balance": portfolio.get("total_value", self.initial_cash),
+            "equity": portfolio.get("equity", []),
+            "baseline": portfolio.get("baseline", []),
+            "baseline_vw": portfolio.get("baseline_vw", []),
+            "momentum": portfolio.get("momentum", []),
+            "equity_return": portfolio.get("equity_return", []),
+            "baseline_return": portfolio.get("baseline_return", []),
+            "baseline_vw_return": portfolio.get("baseline_vw_return", []),
+            "momentum_return": portfolio.get("momentum_return", []),
+        }
+
+        return {
+            "summary": summary,
+            "holdings": holdings,
+            "stats": stats,
+            "trades": trades,
+            "leaderboard": leaderboard,
+        }
+
     def check_file_updates(self) -> Dict[str, bool]:
         """
         Check which dashboard files have been updated since last check
@@ -297,7 +333,7 @@ class StorageService:
     def initialize_empty_dashboard(self):
         """Initialize empty dashboard files with default values"""
         # Summary
-        self.save_file(
+        self.save_export_file(
             "summary",
             {
                 "totalAssetValue": self.initial_cash,
@@ -315,10 +351,10 @@ class StorageService:
         )
 
         # Holdings
-        self.save_file("holdings", [])
+        self.save_export_file("holdings", [])
 
         # Stats
-        self.save_file(
+        self.save_export_file(
             "stats",
             {
                 "totalAssetValue": self.initial_cash,
@@ -335,7 +371,7 @@ class StorageService:
         )
 
         # Trades
-        self.save_file("trades", [])
+        self.save_export_file("trades", [])
 
         # Leaderboard with model info
         self.generate_leaderboard()
@@ -375,7 +411,7 @@ class StorageService:
                 ranking_entries.append(entry)
 
         leaderboard = team_entries + ranking_entries
-        self.save_file("leaderboard", leaderboard)
+        self.save_export_file("leaderboard", leaderboard)
         logger.info("Leaderboard generated with model info")
 
     def update_leaderboard_model_info(self):
@@ -398,7 +434,7 @@ class StorageService:
                 entry["modelName"] = model_name
                 entry["modelProvider"] = model_provider
 
-        self.save_file("leaderboard", existing)
+        self.save_export_file("leaderboard", existing)
         logger.info("Leaderboard model info updated")
 
     def get_current_timestamp_ms(self, date: str = None) -> int:
@@ -653,7 +689,7 @@ class StorageService:
             "momentum": state.get("momentum_history", []),
         }
 
-        self.save_file("summary", summary)
+        self.save_export_file("summary", summary)
 
     def _generate_holdings(
         self,
@@ -715,7 +751,7 @@ class StorageService:
         # Sort by weight
         holdings.sort(key=lambda x: abs(x["weight"]), reverse=True)
 
-        self.save_file("holdings", holdings)
+        self.save_export_file("holdings", holdings)
 
     def _generate_stats(self, state: Dict[str, Any], net_value: float):
         """Generate stats.json"""
@@ -738,7 +774,7 @@ class StorageService:
             },
         }
 
-        self.save_file("stats", stats)
+        self.save_export_file("stats", stats)
 
     def _generate_trades(self, state: Dict[str, Any]):
         """Generate trades.json"""
@@ -764,7 +800,7 @@ class StorageService:
                 },
             )
 
-        self.save_file("trades", trades)
+        self.save_export_file("trades", trades)
 
     # Server State Management Methods
 
@@ -1001,12 +1037,12 @@ class StorageService:
         Args:
             state: Server state dictionary to update
         """
-        # Load dashboard data
+        dashboard_snapshot = self.build_dashboard_snapshot_from_state(state)
-        summary = self.load_file("summary") or {}
+        summary = dashboard_snapshot.get("summary") or {}
-        holdings = self.load_file("holdings") or []
+        holdings = dashboard_snapshot.get("holdings") or []
-        stats = self.load_file("stats") or self._get_default_stats()
+        stats = dashboard_snapshot.get("stats") or self._get_default_stats()
-        trades = self.load_file("trades") or []
+        trades = dashboard_snapshot.get("trades") or []
-        leaderboard = self.load_file("leaderboard") or []
+        leaderboard = dashboard_snapshot.get("leaderboard") or []
         internal_state = self.load_internal_state()
 
         # Update state
@@ -1040,7 +1076,6 @@ class StorageService:
         Start tracking live returns for current trading session.
         Captures current values as session start baseline.
         """
-        summary = self.load_file("summary") or {}
         state = self.load_internal_state()
 
         # Capture current values as session start
@@ -1052,7 +1087,7 @@ class StorageService:
         self._session_start_equity = (
             equity_history[-1]["v"]
             if equity_history
-            else summary.get("totalAssetValue", self.initial_cash)
+            else self.initial_cash
         )
         self._session_start_baseline = (
             baseline_history[-1]["v"]

backend/tests/test_agent_service_app.py

@@ -6,6 +6,7 @@ from pathlib import Path
 from fastapi.testclient import TestClient
 
 from backend.apps.agent_service import create_app
+from backend.api import agents as agents_module
 
 
 def test_agent_service_routes_include_control_plane_endpoints(tmp_path):
@@ -25,3 +26,79 @@ def test_agent_service_excludes_runtime_routes(tmp_path):
 
     assert "/api/runtime/start" not in paths
     assert "/api/runtime/gateway/port" not in paths
+
+
+def test_agent_service_read_routes(monkeypatch, tmp_path):
+    class _FakeSkillsManager:
+        project_root = tmp_path
+
+        def get_agent_asset_dir(self, config_name, agent_id):
+            return tmp_path / "runs" / config_name / "agents" / agent_id
+
+        def resolve_agent_skill_names(self, config_name, agent_id, default_skills=None):
+            return ["demo_skill"]
+
+        def list_agent_skill_catalog(self, config_name, agent_id):
+            return [
+                type(
+                    "Skill",
+                    (),
+                    {
+                        "skill_name": "demo_skill",
+                        "name": "Demo Skill",
+                        "description": "demo",
+                        "version": "1.0.0",
+                        "source": "builtin",
+                        "tools": [],
+                    },
+                )()
+            ]
+
+        def load_agent_skill_document(self, config_name, agent_id, skill_name):
+            return {"skill_name": skill_name, "content": "# demo"}
+
+    class _FakeWorkspaceManager:
+        def load_agent_file(self, config_name, agent_id, filename):
+            return f"{config_name}:{agent_id}:{filename}"
+
+    monkeypatch.setattr(agents_module, "load_agent_profiles", lambda: {"portfolio_manager": {"skills": ["demo_skill"]}})
+    monkeypatch.setattr(agents_module, "get_agent_model_info", lambda agent_id: ("deepseek-v3.2", "DASHSCOPE"))
+    monkeypatch.setattr(
+        agents_module,
+        "load_agent_workspace_config",
+        lambda path: type(
+            "Cfg",
+            (),
+            {
+                "active_tool_groups": ["portfolio_ops"],
+                "disabled_tool_groups": [],
+                "enabled_skills": [],
+                "disabled_skills": [],
+                "prompt_files": ["SOUL.md", "MEMORY.md"],
+            },
+        )(),
+    )
+    monkeypatch.setattr(
+        agents_module,
+        "get_bootstrap_config_for_run",
+        lambda project_root, config_name: type("Bootstrap", (), {"agent_override": lambda self, agent_id: {}})(),
+    )
+
+    app = create_app(project_root=tmp_path)
+    app.dependency_overrides[agents_module.get_skills_manager] = lambda: _FakeSkillsManager()
+    app.dependency_overrides[agents_module.get_workspace_manager] = lambda: _FakeWorkspaceManager()
+
+    with TestClient(app) as client:
+        profile = client.get("/api/workspaces/demo/agents/portfolio_manager/profile")
+        skills = client.get("/api/workspaces/demo/agents/portfolio_manager/skills")
+        detail = client.get("/api/workspaces/demo/agents/portfolio_manager/skills/demo_skill")
+        workspace_file = client.get("/api/workspaces/demo/agents/portfolio_manager/files/MEMORY.md")
+
+    assert profile.status_code == 200
+    assert profile.json()["profile"]["model_name"] == "deepseek-v3.2"
+    assert skills.status_code == 200
+    assert skills.json()["skills"][0]["skill_name"] == "demo_skill"
+    assert detail.status_code == 200
+    assert detail.json()["skill"]["content"] == "# demo"
+    assert workspace_file.status_code == 200
+    assert workspace_file.json()["content"] == "demo:portfolio_manager:MEMORY.md"

backend/tests/test_agent_workspace.py

@@ -13,7 +13,7 @@ class _DummyToolkit:
         return ""
 
 
-def test_workspace_manager_creates_extended_agent_files(tmp_path):
+def test_workspace_manager_creates_core_agent_files(tmp_path):
     manager = WorkspaceManager(project_root=tmp_path)
 
     manager.initialize_default_assets(
@@ -27,7 +27,7 @@ def test_workspace_manager_creates_extended_agent_files(tmp_path):
     assert (asset_dir / "PROFILE.md").exists()
     assert (asset_dir / "AGENTS.md").exists()
     assert (asset_dir / "MEMORY.md").exists()
-    assert (asset_dir / "HEARTBEAT.md").exists()
+    assert (asset_dir / "POLICY.md").exists()
     assert (asset_dir / "agent.yaml").exists()
     assert (asset_dir / "skills" / "installed").is_dir()
     assert (asset_dir / "skills" / "active").is_dir()
@@ -35,6 +35,22 @@ def test_workspace_manager_creates_extended_agent_files(tmp_path):
     assert (asset_dir / "skills" / "local").is_dir()
 
 
+def test_workspace_manager_seeds_risk_prompt_content(tmp_path):
+    manager = WorkspaceManager(project_root=tmp_path)
+    manager.initialize_default_assets(
+        config_name="demo",
+        agent_ids=["risk_manager"],
+        analyst_personas={},
+    )
+
+    asset_dir = tmp_path / "runs" / "demo" / "agents" / "risk_manager"
+    soul = (asset_dir / "SOUL.md").read_text(encoding="utf-8")
+    guide = (asset_dir / "AGENTS.md").read_text(encoding="utf-8")
+
+    assert "风险管理经理" in soul
+    assert "优先使用可用的风险工具量化集中度" in guide
+
+
 def test_agent_workspace_config_controls_prompt_files(tmp_path, monkeypatch):
     manager = WorkspaceManager(project_root=tmp_path)
     manager.initialize_default_assets(
@@ -72,6 +88,32 @@ def test_agent_workspace_config_controls_prompt_files(tmp_path, monkeypatch):
     assert "profile-line" not in prompt
 
 
+def test_prompt_is_built_from_workspace_defaults_without_system_templates(tmp_path, monkeypatch):
+    manager = WorkspaceManager(project_root=tmp_path)
+    manager.initialize_default_assets(
+        config_name="demo",
+        agent_ids=["portfolio_manager"],
+        analyst_personas={},
+    )
+
+    from backend.agents import prompt_factory
+
+    monkeypatch.setattr(
+        prompt_factory,
+        "SkillsManager",
+        lambda: SkillsManager(project_root=tmp_path),
+    )
+
+    prompt = build_agent_system_prompt(
+        agent_id="portfolio_manager",
+        config_name="demo",
+        toolkit=_DummyToolkit(),
+    )
+
+    assert "投资组合经理" in prompt
+    assert "使用 `make_decision` 工具记录每个股票的最终决策" in prompt
+
+
 def test_skills_manager_applies_agent_level_skill_toggles(tmp_path):
     builtin_root = tmp_path / "backend" / "skills" / "builtin"
     for skill_name in ("risk_review", "extra_guard"):

backend/tests/test_agents.py

@@ -311,6 +311,17 @@ class TestRiskAgent:
 
 
 class TestStorageService:
+    def test_storage_service_defaults_to_live_config(self):
+        from backend.services.storage import StorageService
+
+        with tempfile.TemporaryDirectory() as tmpdir:
+            storage = StorageService(
+                dashboard_dir=Path(tmpdir),
+                initial_cash=100000.0,
+            )
+
+            assert storage.config_name == "live"
+
     def test_calculate_portfolio_value_cash_only(self):
         from backend.services.storage import StorageService
 

backend/tests/test_cli.py

@@ -34,7 +34,6 @@ def test_live_runs_incremental_market_store_update_before_start(monkeypatch, tmp
     monkeypatch.setattr(cli.subprocess, "run", fake_run)
 
     cli.live(
-        mock=False,
         config_name="smoke_fullstack",
         host="0.0.0.0",
         port=8765,

backend/tests/test_gateway_support_modules.py

@@ -8,24 +8,6 @@ import pytest
 from backend.services import gateway_cycle_support, gateway_runtime_support
 
 
-class _DummyDashboard:
-    def __init__(self):
-        self.updated = []
-        self.tickers = []
-        self.initial_cash = None
-        self.enable_memory = False
-        self.days_total = 0
-
-    def update(self, **kwargs):
-        self.updated.append(kwargs)
-
-    def stop(self):
-        return None
-
-    def print_final_summary(self):
-        return None
-
-
 class _DummyScheduler:
     def __init__(self):
         self.calls = []
@@ -77,6 +59,15 @@ class _DummyStorage:
             return {"totalAssetValue": self.initial_cash}
         return []
 
+    def build_dashboard_snapshot_from_state(self, state):
+        return {
+            "summary": {"totalAssetValue": self.initial_cash},
+            "holdings": [],
+            "stats": {},
+            "trades": [],
+            "leaderboard": [],
+        }
+
 
 class _DummyPM:
     def __init__(self):
@@ -119,7 +110,6 @@ def make_gateway_stub():
         },
         storage=_DummyStorage(),
         state_sync=_DummyStateSync(),
-        _dashboard=_DummyDashboard(),
         _watchlist_ingest_task=None,
         _market_status_task=None,
         _backtest_task=None,

backend/tests/test_heartbeat_hook.py

@@ -1,69 +0,0 @@
-# -*- coding: utf-8 -*-
-"""Tests for HeartbeatHook."""
-import tempfile
-from pathlib import Path
-
-import pytest
-
-from backend.agents.base.hooks import HeartbeatHook
-
-
-class TestHeartbeatHook:
-    """Tests for HeartbeatHook._read_heartbeat_content."""
-
-    def test_read_heartbeat_content_with_content(self, tmp_path):
-        """Test reading HEARTBEAT.md when it exists and has content."""
-        ws_dir = tmp_path / "analyst_workspace"
-        ws_dir.mkdir()
-        hb_file = ws_dir / "HEARTBEAT.md"
-        hb_file.write_text("# 定期主动检查\n\n- [ ] 持仓是否健康\n", encoding="utf-8")
-
-        hook = HeartbeatHook(workspace_dir=ws_dir)
-        content = hook._read_heartbeat_content()
-
-        assert content is not None
-        assert "# 定期主动检查" in content
-        assert "持仓是否健康" in content
-
-    def test_read_heartbeat_content_absent(self, tmp_path):
-        """Test reading when HEARTBEAT.md does not exist."""
-        ws_dir = tmp_path / "analyst_workspace"
-        ws_dir.mkdir()
-
-        hook = HeartbeatHook(workspace_dir=ws_dir)
-        content = hook._read_heartbeat_content()
-
-        assert content is None
-
-    def test_read_heartbeat_content_empty(self, tmp_path):
-        """Test reading when HEARTBEAT.md is empty."""
-        ws_dir = tmp_path / "analyst_workspace"
-        ws_dir.mkdir()
-        hb_file = ws_dir / "HEARTBEAT.md"
-        hb_file.write_text("", encoding="utf-8")
-
-        hook = HeartbeatHook(workspace_dir=ws_dir)
-        content = hook._read_heartbeat_content()
-
-        assert content is None
-
-    def test_read_heartbeat_content_whitespace_only(self, tmp_path):
-        """Test reading when HEARTBEAT.md contains only whitespace."""
-        ws_dir = tmp_path / "analyst_workspace"
-        ws_dir.mkdir()
-        hb_file = ws_dir / "HEARTBEAT.md"
-        hb_file.write_text("   \n\n  ", encoding="utf-8")
-
-        hook = HeartbeatHook(workspace_dir=ws_dir)
-        content = hook._read_heartbeat_content()
-
-        assert content is None
-
-    def test_completed_flag_path(self, tmp_path):
-        """Test that completion flag is placed in workspace directory."""
-        ws_dir = tmp_path / "analyst_workspace"
-        ws_dir.mkdir()
-
-        hook = HeartbeatHook(workspace_dir=ws_dir)
-
-        assert hook._completed_flag == ws_dir / ".heartbeat_completed"

backend/tests/test_market_ingest.py

@@ -0,0 +1,81 @@
+# -*- coding: utf-8 -*-
+"""Tests for market ingest watermark handling."""
+
+from backend.data import market_ingest
+
+
+class _FakeStore:
+    def __init__(self, *, last_news_fetch=None, latest_news_date=None):
+        self._watermarks = {
+            "symbol": "AAPL",
+            "last_price_fetch": None,
+            "last_news_fetch": last_news_fetch,
+        }
+        self._latest_news_date = latest_news_date
+        self.updated = []
+
+    def get_ticker_watermarks(self, symbol):
+        return dict(self._watermarks)
+
+    def get_latest_news_date(self, symbol):
+        return self._latest_news_date
+
+    def upsert_ticker(self, **kwargs):
+        return None
+
+    def upsert_ohlc(self, symbol, rows, source="polygon"):
+        return len(rows)
+
+    def upsert_news(self, symbol, rows, source="polygon"):
+        return len(rows)
+
+    def update_fetch_watermark(self, **kwargs):
+        self.updated.append(kwargs)
+
+
+def test_refresh_news_incremental_does_not_advance_watermark_without_news(monkeypatch):
+    store = _FakeStore(last_news_fetch="2026-03-28", latest_news_date="2026-03-28")
+
+    monkeypatch.setattr(market_ingest, "fetch_ticker_details", lambda ticker: {"name": ticker, "sic_description": None, "active": True})
+
+    class _Router:
+        def get_company_news(self, **kwargs):
+            return [], "polygon"
+
+    monkeypatch.setattr(market_ingest, "DataProviderRouter", lambda: _Router())
+    monkeypatch.setattr(market_ingest, "align_news_for_symbol", lambda store, ticker: 0)
+
+    result = market_ingest.refresh_news_incremental(
+        "AAPL",
+        end_date="2026-03-29",
+        store=store,
+    )
+
+    assert result["start_news_date"] == "2026-03-29"
+    assert result["news"] == 0
+    assert store.updated[-1]["news_date"] is None
+
+
+def test_refresh_news_incremental_clamps_future_watermark_to_latest_stored_date(monkeypatch):
+    store = _FakeStore(last_news_fetch="2026-03-30", latest_news_date="2026-03-28")
+    captured = {}
+
+    monkeypatch.setattr(market_ingest, "fetch_ticker_details", lambda ticker: {"name": ticker, "sic_description": None, "active": True})
+
+    class _Router:
+        def get_company_news(self, **kwargs):
+            captured.update(kwargs)
+            return [], "polygon"
+
+    monkeypatch.setattr(market_ingest, "DataProviderRouter", lambda: _Router())
+    monkeypatch.setattr(market_ingest, "align_news_for_symbol", lambda store, ticker: 0)
+
+    result = market_ingest.refresh_news_incremental(
+        "AAPL",
+        end_date="2026-03-29",
+        store=store,
+    )
+
+    assert result["start_news_date"] == "2026-03-29"
+    assert captured["start_date"] == "2026-03-29"
+    assert captured["end_date"] == "2026-03-29"

backend/tests/test_market_service.py

@@ -2,153 +2,12 @@
 # pylint: disable=W0212
 import asyncio
 import time
+import logging
 from unittest.mock import MagicMock, AsyncMock, patch
 import pytest
 from backend.services.market import MarketService
-from backend.data.mock_price_manager import MockPriceManager
 from backend.data.polling_price_manager import PollingPriceManager
-
+from backend.llm.models import RetryChatModel
-
-class TestMockPriceManager:
-    def test_init_default(self):
-        manager = MockPriceManager()
-
-        assert manager.poll_interval == 10
-        assert manager.volatility == 0.5
-        assert manager.running is False
-        assert len(manager.subscribed_symbols) == 0
-
-    def test_init_custom(self):
-        manager = MockPriceManager(poll_interval=5, volatility=1.0)
-
-        assert manager.poll_interval == 5
-        assert manager.volatility == 1.0
-
-    def test_subscribe(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL", "MSFT"])
-
-        assert "AAPL" in manager.subscribed_symbols
-        assert "MSFT" in manager.subscribed_symbols
-        assert manager.base_prices["AAPL"] == 237.50  # default price
-        assert manager.base_prices["MSFT"] == 425.30  # default price
-
-    def test_subscribe_with_base_prices(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        assert manager.base_prices["AAPL"] == 100.0
-        assert manager.open_prices["AAPL"] == 100.0
-        assert manager.latest_prices["AAPL"] == 100.0
-
-    def test_subscribe_unknown_symbol(self):
-        manager = MockPriceManager()
-        manager.subscribe(["UNKNOWN"])
-
-        assert "UNKNOWN" in manager.subscribed_symbols
-        assert manager.base_prices["UNKNOWN"] > 0  # random price generated
-
-    def test_unsubscribe(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL", "MSFT"])
-        manager.unsubscribe(["AAPL"])
-
-        assert "AAPL" not in manager.subscribed_symbols
-        assert "MSFT" in manager.subscribed_symbols
-
-    def test_add_price_callback(self):
-        manager = MockPriceManager()
-        callback = MagicMock()
-        manager.add_price_callback(callback)
-
-        assert callback in manager.price_callbacks
-
-    def test_generate_price_update_within_bounds(self):
-        manager = MockPriceManager(volatility=0.5)
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        for _ in range(100):
-            new_price = manager._generate_price_update("AAPL")
-            # Should be within +/-10% of open
-            assert 90.0 <= new_price <= 110.0
-
-    def test_update_prices_triggers_callback(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        callback = MagicMock()
-        manager.add_price_callback(callback)
-
-        manager._update_prices()
-
-        callback.assert_called_once()
-        call_args = callback.call_args[0][0]
-        assert call_args["symbol"] == "AAPL"
-        assert "price" in call_args
-        assert "timestamp" in call_args
-
-    def test_start_stop(self):
-        manager = MockPriceManager(poll_interval=1)
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        manager.start()
-        assert manager.running is True
-
-        time.sleep(0.1)  # let thread start
-
-        manager.stop()
-        assert manager.running is False
-
-    def test_start_without_subscription(self):
-        manager = MockPriceManager()
-        manager.start()
-
-        assert (
-            manager.running is False
-        )  # should not start without subscriptions
-
-    def test_get_latest_price(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        price = manager.get_latest_price("AAPL")
-        assert price == 100.0
-
-    def test_get_latest_price_unknown(self):
-        manager = MockPriceManager()
-        price = manager.get_latest_price("UNKNOWN")
-        assert price is None
-
-    def test_get_all_latest_prices(self):
-        manager = MockPriceManager()
-        manager.subscribe(
-            ["AAPL", "MSFT"],
-            base_prices={"AAPL": 100.0, "MSFT": 200.0},
-        )
-
-        prices = manager.get_all_latest_prices()
-        assert prices["AAPL"] == 100.0
-        assert prices["MSFT"] == 200.0
-
-    def test_reset_open_prices(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-        manager.latest_prices["AAPL"] = 105.0
-
-        manager.reset_open_prices()
-
-        # Open price should change (based on latest with small gap)
-        assert manager.open_prices["AAPL"] != 100.0
-
-    def test_set_base_price(self):
-        manager = MockPriceManager()
-        manager.subscribe(["AAPL"], base_prices={"AAPL": 100.0})
-
-        manager.set_base_price("AAPL", 150.0)
-
-        assert manager.base_prices["AAPL"] == 150.0
-        assert manager.open_prices["AAPL"] == 150.0
-        assert manager.latest_prices["AAPL"] == 150.0
 
 
 class TestPollingPriceManager:
@@ -231,37 +90,67 @@ class TestPollingPriceManager:
 
         assert len(manager.open_prices) == 0
 
+    def test_fetch_prices_suppresses_repeated_failures(self, caplog):
+        manager = PollingPriceManager(provider="yfinance", poll_interval=10)
+        manager.subscribe(["AAPL"])
+
+        with patch.object(manager, "_fetch_quote", side_effect=ValueError("empty quote")):
+            with caplog.at_level(logging.DEBUG):
+                for _ in range(3):
+                    manager._fetch_prices()
+
+        assert manager._failure_counts["AAPL"] == 3
+        warning_messages = [record.message for record in caplog.records if record.levelno >= logging.WARNING]
+        assert any("Failed to fetch AAPL price: empty quote" in message for message in warning_messages)
+
+    def test_fetch_prices_logs_recovery_after_failure(self, caplog):
+        manager = PollingPriceManager(provider="yfinance", poll_interval=10)
+        manager.subscribe(["AAPL"])
+
+        with patch.object(
+            manager,
+            "_fetch_quote",
+            side_effect=[
+                ValueError("temporary outage"),
+                {"c": 100.0, "o": 99.0, "h": 101.0, "l": 98.0, "pc": 99.5, "d": 0.5, "dp": 0.5, "t": 1},
+            ],
+        ):
+            with caplog.at_level(logging.INFO):
+                manager._fetch_prices()
+                manager._fetch_prices()
+
+        assert "AAPL" not in manager._failure_counts
+        assert any("recovered after 1 consecutive failures" in record.message for record in caplog.records)
+
+
+class TestRetryChatModel:
+    @pytest.mark.asyncio
+    async def test_async_retry_recovers_from_disconnect(self):
+        attempts = {"count": 0}
+
+        class FakeAsyncModel:
+            model_name = "fake-async-model"
+
+            async def __call__(self, *args, **kwargs):
+                attempts["count"] += 1
+                if attempts["count"] < 2:
+                    raise RuntimeError("Server disconnected")
+                return {"ok": True}
+
+        wrapped = RetryChatModel(FakeAsyncModel(), max_retries=2, initial_delay=0.01)
+        result = await wrapped("hello")
+
+        assert result == {"ok": True}
+        assert attempts["count"] == 2
+
 
 class TestMarketService:
-    def test_init_mock_mode(self):
+    @patch("backend.services.market.get_data_sources", return_value=["yfinance", "local_csv"])
-        service = MarketService(
-            tickers=["AAPL", "MSFT"],
-            poll_interval=10,
-            mock_mode=True,
-        )
-
-        assert service.tickers == ["AAPL", "MSFT"]
-        assert service.poll_interval == 10
-        assert service.mock_mode is True
-        assert service.running is False
-
-    def test_init_real_mode(self):
-        service = MarketService(
-            tickers=["AAPL"],
-            mock_mode=False,
-            api_key="test_key",
-        )
-
-        assert service.mock_mode is False
-        assert service.api_key == "test_key"
-
-    @patch("backend.services.market.get_data_source", return_value="yfinance")
     @patch.object(PollingPriceManager, "start")
-    def test_start_real_mode_with_yfinance(self, _mock_start, _mock_source):
+    def test_start_real_mode_with_yfinance(self, _mock_start, _mock_sources):
         service = MarketService(
             tickers=["AAPL"],
             poll_interval=10,
-            mock_mode=False,
         )
 
         service._start_real_mode()
@@ -269,30 +158,24 @@ class TestMarketService:
         assert isinstance(service._price_manager, PollingPriceManager)
         assert service._price_manager.provider == "yfinance"
 
-    @pytest.mark.asyncio
+    @patch("backend.services.market.get_data_sources", return_value=["financial_datasets", "yfinance", "local_csv"])
-    async def test_start_mock_mode(self):
+    @patch.object(PollingPriceManager, "start")
+    def test_start_real_mode_uses_first_supported_live_provider(self, _mock_start, _mock_sources):
         service = MarketService(
             tickers=["AAPL"],
             poll_interval=10,
-            mock_mode=True,
         )
 
-        broadcast_func = AsyncMock()
+        service._start_real_mode()
 
-        await service.start(broadcast_func)
+        assert isinstance(service._price_manager, PollingPriceManager)
+        assert service._price_manager.provider == "yfinance"
 
-        assert service.running is True
+    @patch("backend.services.market.get_data_sources", return_value=["finnhub", "yfinance"])
-        assert service._price_manager is not None
-        assert isinstance(service._price_manager, MockPriceManager)
-
-        service.stop()
-
-    @patch("backend.services.market.get_data_source", return_value="finnhub")
     @pytest.mark.asyncio
-    async def test_start_real_mode_without_api_key(self, _mock_source):
+    async def test_start_real_mode_without_api_key(self, _mock_sources):
         service = MarketService(
             tickers=["AAPL"],
-            mock_mode=False,
             api_key=None,
         )
 
@@ -307,11 +190,12 @@ class TestMarketService:
     async def test_start_already_running(self):
         service = MarketService(
             tickers=["AAPL"],
-            mock_mode=True,
+            backtest_mode=True,
         )
 
         broadcast_func = AsyncMock()
 
+        # First start with backtest mode
         await service.start(broadcast_func)
         assert service.running is True
 
@@ -323,7 +207,7 @@ class TestMarketService:
     def test_stop(self):
         service = MarketService(
             tickers=["AAPL"],
-            mock_mode=True,
+            backtest_mode=True,
         )
         service.running = True
         service._price_manager = MagicMock()
@@ -336,7 +220,7 @@ class TestMarketService:
     def test_stop_when_not_running(self):
         service = MarketService(
             tickers=["AAPL"],
-            mock_mode=True,
+            backtest_mode=True,
         )
 
         # Should not raise
@@ -344,20 +228,20 @@ class TestMarketService:
         assert service.running is False
 
     def test_get_price_sync(self):
-        service = MarketService(tickers=["AAPL"], mock_mode=True)
+        service = MarketService(tickers=["AAPL"], backtest_mode=True)
         service.cache["AAPL"] = {"price": 150.0, "open": 148.0}
 
         price = service.get_price_sync("AAPL")
         assert price == 150.0
 
     def test_get_price_sync_not_found(self):
-        service = MarketService(tickers=["AAPL"], mock_mode=True)
+        service = MarketService(tickers=["AAPL"], backtest_mode=True)
 
         price = service.get_price_sync("MSFT")
         assert price is None
 
     def test_get_all_prices(self):
-        service = MarketService(tickers=["AAPL", "MSFT"], mock_mode=True)
+        service = MarketService(tickers=["AAPL", "MSFT"], backtest_mode=True)
         service.cache["AAPL"] = {"price": 150.0}
         service.cache["MSFT"] = {"price": 400.0}
 
@@ -368,7 +252,7 @@ class TestMarketService:
 
     @pytest.mark.asyncio
     async def test_broadcast_price_update(self):
-        service = MarketService(tickers=["AAPL"], mock_mode=True)
+        service = MarketService(tickers=["AAPL"], backtest_mode=True)
         service._broadcast_func = AsyncMock()
 
         price_data = {
@@ -388,7 +272,7 @@ class TestMarketService:
 
     @pytest.mark.asyncio
     async def test_broadcast_price_update_no_func(self):
-        service = MarketService(tickers=["AAPL"], mock_mode=True)
+        service = MarketService(tickers=["AAPL"], backtest_mode=True)
         service._broadcast_func = None
 
         price_data = {"symbol": "AAPL", "price": 150.0, "open": 148.0}
@@ -396,67 +280,6 @@ class TestMarketService:
         # Should not raise
         await service._broadcast_price_update(price_data)
 
-    @pytest.mark.asyncio
-    async def test_price_callback_thread_safety(self):
-        service = MarketService(
-            tickers=["AAPL"],
-            poll_interval=1,
-            mock_mode=True,
-        )
-
-        received_prices = []
-
-        async def capture_broadcast(msg):
-            received_prices.append(msg)
-
-        await service.start(capture_broadcast)
-
-        # Wait for at least one price update
-        await asyncio.sleep(1.5)
-
-        service.stop()
-
-        # Should have received at least one price update
-        assert len(received_prices) >= 1
-        assert received_prices[0]["type"] == "price_update"
-
-
-class TestMarketServiceIntegration:
-    @pytest.mark.asyncio
-    async def test_full_mock_cycle(self):
-        service = MarketService(
-            tickers=["AAPL", "MSFT"],
-            poll_interval=1,
-            mock_mode=True,
-        )
-
-        messages = []
-
-        async def collect_messages(msg):
-            messages.append(msg)
-
-        await service.start(collect_messages)
-
-        # Wait for price updates
-        await asyncio.sleep(2.5)
-
-        service.stop()
-
-        # Should have received multiple price updates
-        assert len(messages) >= 2
-
-        # Check message structure
-        symbols_seen = set()
-        for msg in messages:
-            assert msg["type"] == "price_update"
-            assert "symbol" in msg
-            assert "price" in msg
-            assert "ret" in msg
-            symbols_seen.add(msg["symbol"])
-
-        # Should have prices for both tickers
-        assert "AAPL" in symbols_seen or "MSFT" in symbols_seen
-
 
 if __name__ == "__main__":
     pytest.main([__file__, "-v"])

backend/tests/test_news_domain.py

@@ -80,7 +80,7 @@ def test_get_enriched_news_returns_rows_without_enrichment_when_present(monkeypa
     monkeypatch.setattr(
         news_domain,
         "ensure_news_fresh",
-        lambda store, ticker, target_date=None: {
+        lambda store, ticker, target_date=None, refresh_if_stale=False: {
             "ticker": ticker,
             "target_date": target_date,
             "last_news_fetch": target_date,
@@ -109,7 +109,7 @@ def test_get_story_and_similar_days_delegate(monkeypatch):
     monkeypatch.setattr(
         news_domain,
         "ensure_news_fresh",
-        lambda store, ticker, target_date=None: {
+        lambda store, ticker, target_date=None, refresh_if_stale=False: {
             "ticker": ticker,
             "target_date": target_date,
             "last_news_fetch": target_date,
@@ -137,12 +137,38 @@ def test_get_story_and_similar_days_delegate(monkeypatch):
     assert "freshness" in similar
 
 
+def test_get_enriched_news_defaults_to_read_only_freshness(monkeypatch):
+    store = _FakeStore()
+    ensure_calls = []
+
+    def fake_ensure(store, ticker, target_date=None, refresh_if_stale=False):
+        ensure_calls.append(refresh_if_stale)
+        return {
+            "ticker": ticker,
+            "target_date": target_date,
+            "last_news_fetch": target_date,
+            "refreshed": False,
+        }
+
+    monkeypatch.setattr(news_domain, "ensure_news_fresh", fake_ensure)
+    monkeypatch.setattr(news_domain, "news_rows_need_enrichment", lambda rows: False)
+
+    payload = news_domain.get_enriched_news(
+        store,
+        ticker="AAPL",
+        end_date="2026-03-16",
+    )
+
+    assert payload["ticker"] == "AAPL"
+    assert ensure_calls == [False]
+
+
 def test_get_range_explain_payload_uses_article_ids(monkeypatch):
     store = _FakeStore()
     monkeypatch.setattr(
         news_domain,
         "ensure_news_fresh",
-        lambda store, ticker, target_date=None: {
+        lambda store, ticker, target_date=None, refresh_if_stale=False: {
             "ticker": ticker,
             "target_date": target_date,
             "last_news_fetch": target_date,

backend/tests/test_openclaw_cli_service.py

@@ -0,0 +1,60 @@
+# -*- coding: utf-8 -*-
+"""Tests for the OpenClaw CLI service wrapper."""
+
+from pathlib import Path
+
+import pytest
+
+from backend.services.openclaw_cli import OpenClawCliError, OpenClawCliService
+
+
+class _Completed:
+    def __init__(self, *, returncode=0, stdout="", stderr=""):
+        self.returncode = returncode
+        self.stdout = stdout
+        self.stderr = stderr
+
+
+def test_openclaw_cli_service_runs_json_command(monkeypatch, tmp_path):
+    captured = {}
+
+    def _fake_run(command, **kwargs):
+        captured["command"] = command
+        captured["cwd"] = kwargs["cwd"]
+        return _Completed(stdout='{"sessions":[{"key":"main/session-1"}]}')
+
+    monkeypatch.setattr("backend.services.openclaw_cli.subprocess.run", _fake_run)
+
+    service = OpenClawCliService(base_command=["openclaw"], cwd=tmp_path, timeout_seconds=3)
+    payload = service.list_sessions()
+
+    assert payload["sessions"][0]["key"] == "main/session-1"
+    assert captured["command"] == ["openclaw", "sessions", "--json"]
+    assert captured["cwd"] == tmp_path
+
+
+def test_openclaw_cli_service_raises_on_failure(monkeypatch, tmp_path):
+    def _fake_run(command, **kwargs):
+        return _Completed(returncode=7, stdout="", stderr="boom")
+
+    monkeypatch.setattr("backend.services.openclaw_cli.subprocess.run", _fake_run)
+
+    service = OpenClawCliService(base_command=["openclaw"], cwd=tmp_path, timeout_seconds=3)
+
+    with pytest.raises(OpenClawCliError) as exc_info:
+        service.list_cron_jobs()
+
+    assert exc_info.value.exit_code == 7
+    assert exc_info.value.stderr == "boom"
+
+
+def test_openclaw_cli_service_can_extract_single_session(monkeypatch, tmp_path):
+    def _fake_run(command, **kwargs):
+        return _Completed(stdout='{"sessions":[{"key":"main/session-1","agentId":"main"}]}')
+
+    monkeypatch.setattr("backend.services.openclaw_cli.subprocess.run", _fake_run)
+
+    service = OpenClawCliService(base_command=["openclaw"], cwd=tmp_path, timeout_seconds=3)
+    session = service.get_session("main/session-1")
+
+    assert session["agentId"] == "main"

backend/tests/test_openclaw_service_app.py

@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+"""Tests for the extracted OpenClaw service app surface."""
+
+from fastapi.testclient import TestClient
+
+from backend.apps.openclaw_service import create_app
+from backend.api import openclaw as openclaw_module
+
+
+class _FakeOpenClawCliService:
+    def health(self):
+        return {
+            "status": "healthy",
+            "service": "openclaw-service",
+            "base_command": ["openclaw"],
+            "cwd": "/tmp/openclaw",
+            "binary_resolved": True,
+            "reference_entry_available": True,
+            "timeout_seconds": 15.0,
+        }
+
+    def status(self):
+        return {"runtimeVersion": "2026.3.24"}
+
+    def list_sessions(self):
+        return {
+            "sessions": [
+                {"key": "main/session-1", "agentId": "main"},
+                {"key": "analyst/session-2", "agentId": "analyst"},
+            ]
+        }
+
+    def get_session(self, session_key: str):
+        for session in self.list_sessions()["sessions"]:
+            if session["key"] == session_key:
+                return session
+        raise KeyError(session_key)
+
+    def get_session_history(self, session_key: str, *, limit: int = 20):
+        return {
+            "sessionKey": session_key,
+            "limit": limit,
+            "items": [{"role": "assistant", "text": "hello"}],
+        }
+
+    def list_cron_jobs(self):
+        return {"jobs": [{"id": "job-1", "name": "Daily sync"}]}
+
+    def list_approvals(self):
+        return {"approvals": [{"id": "ap-1", "status": "pending"}]}
+
+
+def test_openclaw_service_routes_are_exposed():
+    app = create_app()
+    paths = {route.path for route in app.routes}
+
+    assert "/health" in paths
+    assert "/api/status" in paths
+    assert "/api/openclaw/status" in paths
+    assert "/api/openclaw/sessions" in paths
+    assert "/api/openclaw/sessions/{session_key:path}" in paths
+    assert "/api/openclaw/sessions/{session_key:path}/history" in paths
+    assert "/api/openclaw/cron" in paths
+    assert "/api/openclaw/approvals" in paths
+
+
+def test_openclaw_service_read_routes():
+    app = create_app()
+    app.dependency_overrides[openclaw_module.get_openclaw_cli_service] = (
+        lambda: _FakeOpenClawCliService()
+    )
+
+    with TestClient(app) as client:
+        health = client.get("/health")
+        status = client.get("/api/status")
+        openclaw_status = client.get("/api/openclaw/status")
+        sessions = client.get("/api/openclaw/sessions")
+        session = client.get("/api/openclaw/sessions/main/session-1")
+        history = client.get("/api/openclaw/sessions/main/session-1/history", params={"limit": 5})
+        cron = client.get("/api/openclaw/cron")
+        approvals = client.get("/api/openclaw/approvals")
+
+    assert health.status_code == 200
+    assert health.json()["service"] == "openclaw-service"
+    assert status.status_code == 200
+    assert status.json()["status"] == "operational"
+    assert openclaw_status.status_code == 200
+    assert openclaw_status.json()["runtimeVersion"] == "2026.3.24"
+    assert sessions.status_code == 200
+    assert len(sessions.json()["sessions"]) == 2
+    assert session.status_code == 200
+    assert session.json()["session"]["agentId"] == "main"
+    assert history.status_code == 200
+    assert history.json()["limit"] == 5
+    assert cron.status_code == 200
+    assert cron.json()["jobs"][0]["id"] == "job-1"
+    assert approvals.status_code == 200
+    assert approvals.json()["approvals"][0]["id"] == "ap-1"
+
+
+def test_openclaw_service_session_404():
+    app = create_app()
+    app.dependency_overrides[openclaw_module.get_openclaw_cli_service] = (
+        lambda: _FakeOpenClawCliService()
+    )
+
+    with TestClient(app) as client:
+        response = client.get("/api/openclaw/sessions/missing")
+
+    assert response.status_code == 404

backend/tests/test_openclaw_websocket_client.py

@@ -0,0 +1,74 @@
+# -*- coding: utf-8 -*-
+"""Tests for the OpenClaw WebSocket client session helpers."""
+
+import pytest
+
+from shared.client.openclaw_websocket_client import OpenClawWebSocketClient
+
+
+@pytest.mark.asyncio
+async def test_resolve_session_parses_gateway_key_response():
+    client = OpenClawWebSocketClient(gateway_token="test-token")
+
+    async def fake_send_request(method, params=None, _allow_handshake=False):
+        assert method == "sessions.resolve"
+        assert params["agentId"] == "main"
+        return {"ok": True, "key": "agent:main:main"}
+
+    client._send_request = fake_send_request  # type: ignore[method-assign]
+
+    resolved = await client.resolve_session(agent_id="main")
+
+    assert resolved == "agent:main:main"
+
+
+@pytest.mark.asyncio
+async def test_send_message_uses_session_send_payload():
+    client = OpenClawWebSocketClient(gateway_token="test-token")
+
+    async def fake_send_request(method, params=None, _allow_handshake=False):
+        assert method == "sessions.send"
+        assert params == {
+            "key": "agent:main:main",
+            "message": "hello",
+            "thinking": "medium",
+        }
+        return {"ok": True, "runId": "run-1"}
+
+    client._send_request = fake_send_request  # type: ignore[method-assign]
+
+    result = await client.send_message("agent:main:main", "hello", thinking="medium")
+
+    assert result["runId"] == "run-1"
+
+
+@pytest.mark.asyncio
+async def test_get_session_history_uses_sessions_preview():
+    client = OpenClawWebSocketClient(gateway_token="test-token")
+
+    async def fake_send_request(method, params=None, _allow_handshake=False):
+        assert method == "sessions.preview"
+        assert params == {"keys": ["agent:main:main"], "limit": 12}
+        return {"previews": []}
+
+    client._send_request = fake_send_request  # type: ignore[method-assign]
+
+    result = await client.get_session_history("agent:main:main", limit=12)
+
+    assert result == {"previews": []}
+
+
+@pytest.mark.asyncio
+async def test_unsubscribe_uses_session_messages_unsubscribe():
+    client = OpenClawWebSocketClient(gateway_token="test-token")
+
+    async def fake_send_request(method, params=None, _allow_handshake=False):
+        assert method == "sessions.messages.unsubscribe"
+        assert params == {"key": "agent:main:main"}
+        return {"subscribed": False}
+
+    client._send_request = fake_send_request  # type: ignore[method-assign]
+
+    result = await client.unsubscribe("agent:main:main")
+
+    assert result == {"subscribed": False}

backend/tests/test_runtime_service_app.py

@@ -2,6 +2,7 @@
 """Tests for the extracted runtime service app surface."""
 
 import json
+from pathlib import Path
 
 from fastapi.testclient import TestClient
 
@@ -17,6 +18,8 @@ def test_runtime_service_routes_are_exposed():
     assert "/api/status" in paths
     assert "/api/runtime/start" in paths
     assert "/api/runtime/stop" in paths
+    assert "/api/runtime/cleanup" in paths
+    assert "/api/runtime/history" in paths
     assert "/api/runtime/current" in paths
     assert "/api/runtime/gateway/port" in paths
 
@@ -192,3 +195,170 @@ def test_runtime_service_update_runtime_config_persists_bootstrap(monkeypatch, t
     assert payload["bootstrap"]["schedule_mode"] == "intraday"
     assert payload["resolved"]["interval_minutes"] == 15
     assert "interval_minutes: 15" in (run_dir / "BOOTSTRAP.md").read_text(encoding="utf-8")
+
+
+def test_prune_old_timestamped_runs_keeps_named_runs(monkeypatch, tmp_path):
+    runs_dir = tmp_path / "runs"
+    runs_dir.mkdir()
+
+    keep_dirs = ["20260324_110000", "20260324_120000"]
+    prune_dir = "20260324_100000"
+    named_dir = "smoke_fullstack"
+
+    for name in [*keep_dirs, prune_dir, named_dir]:
+        (runs_dir / name).mkdir(parents=True)
+
+    monkeypatch.setattr(runtime_module, "PROJECT_ROOT", tmp_path)
+
+    pruned = runtime_module._prune_old_timestamped_runs(keep=1, exclude_run_ids={"20260324_120000"})
+
+    assert prune_dir in pruned
+    assert (runs_dir / named_dir).exists()
+    assert (runs_dir / "20260324_120000").exists()
+    assert (runs_dir / "20260324_110000").exists()
+
+
+def test_runtime_cleanup_endpoint_prunes_old_runs(monkeypatch, tmp_path):
+    runs_dir = tmp_path / "runs"
+    runs_dir.mkdir()
+
+    for name in ["20260324_090000", "20260324_100000", "20260324_110000", "smoke_fullstack"]:
+        (runs_dir / name).mkdir(parents=True)
+
+    monkeypatch.setattr(runtime_module, "PROJECT_ROOT", tmp_path)
+    monkeypatch.setattr(runtime_module, "_is_gateway_running", lambda: False)
+
+    with TestClient(create_app()) as client:
+        response = client.post("/api/runtime/cleanup?keep=1")
+
+    assert response.status_code == 200
+    payload = response.json()
+    assert payload["status"] == "ok"
+    assert sorted(payload["pruned_run_ids"]) == ["20260324_090000", "20260324_100000"]
+    assert (runs_dir / "20260324_110000").exists()
+    assert (runs_dir / "smoke_fullstack").exists()
+
+
+def test_runtime_history_lists_recent_runs(monkeypatch, tmp_path):
+    run_dir = tmp_path / "runs" / "20260324_120000"
+    (run_dir / "state").mkdir(parents=True)
+    (run_dir / "team_dashboard").mkdir(parents=True)
+    (run_dir / "state" / "runtime_state.json").write_text(
+        json.dumps(
+            {
+                "context": {
+                    "config_name": "20260324_120000",
+                    "run_dir": str(run_dir),
+                    "bootstrap_values": {"tickers": ["AAPL"]},
+                },
+                "events": [],
+            }
+        ),
+        encoding="utf-8",
+    )
+    (run_dir / "team_dashboard" / "summary.json").write_text(
+        json.dumps({"totalTrades": 3, "totalAssetValue": 123456.0}),
+        encoding="utf-8",
+    )
+
+    monkeypatch.setattr(runtime_module, "PROJECT_ROOT", tmp_path)
+
+    with TestClient(create_app()) as client:
+        response = client.get("/api/runtime/history?limit=5")
+
+    assert response.status_code == 200
+    payload = response.json()
+    assert payload["runs"][0]["run_id"] == "20260324_120000"
+    assert payload["runs"][0]["total_trades"] == 3
+
+
+def test_restore_run_assets_copies_state(monkeypatch, tmp_path):
+    source_run = tmp_path / "runs" / "20260324_100000"
+    (source_run / "team_dashboard").mkdir(parents=True)
+    (source_run / "state").mkdir(parents=True)
+    (source_run / "agents").mkdir(parents=True)
+    (source_run / "team_dashboard" / "_internal_state.json").write_text("{}", encoding="utf-8")
+    (source_run / "state" / "server_state.json").write_text("{}", encoding="utf-8")
+
+    target_run = tmp_path / "runs" / "20260324_130000"
+
+    monkeypatch.setattr(runtime_module, "PROJECT_ROOT", tmp_path)
+
+    runtime_module._restore_run_assets("20260324_100000", target_run)
+
+    assert (target_run / "team_dashboard" / "_internal_state.json").exists()
+    assert (target_run / "state" / "server_state.json").exists()
+
+
+def test_start_runtime_restore_reuses_historical_run_id(monkeypatch, tmp_path):
+    run_dir = tmp_path / "runs" / "20260324_100000"
+    (run_dir / "state").mkdir(parents=True)
+    (run_dir / "state" / "runtime_state.json").write_text(
+        json.dumps(
+            {
+                "context": {
+                    "config_name": "20260324_100000",
+                    "run_dir": str(run_dir),
+                    "bootstrap_values": {
+                        "tickers": ["AAPL"],
+                        "schedule_mode": "intraday",
+                        "interval_minutes": 30,
+                        "trigger_time": "now",
+                        "max_comm_cycles": 2,
+                        "initial_cash": 100000.0,
+                        "margin_requirement": 0.0,
+                        "enable_memory": False,
+                        "mode": "live",
+                        "poll_interval": 10,
+                    },
+                }
+            }
+        ),
+        encoding="utf-8",
+    )
+
+    class _DummyManager:
+        def __init__(self, config_name, run_dir, bootstrap):
+            self.config_name = config_name
+            self.run_dir = Path(run_dir)
+            self.bootstrap = bootstrap
+            self.context = None
+
+        def prepare_run(self):
+            self.context = type(
+                "Ctx",
+                (),
+                {
+                    "config_name": self.config_name,
+                    "run_dir": self.run_dir,
+                    "bootstrap_values": self.bootstrap,
+                },
+            )()
+            return self.context
+
+    class _DummyProcess:
+        def poll(self):
+            return None
+
+    monkeypatch.setattr(runtime_module, "PROJECT_ROOT", tmp_path)
+    monkeypatch.setattr(runtime_module, "_find_available_port", lambda start_port=8765, max_port=9000: 8765)
+    monkeypatch.setattr(runtime_module, "_start_gateway_process", lambda **kwargs: _DummyProcess())
+    monkeypatch.setattr(runtime_module, "_stop_gateway", lambda: True)
+    monkeypatch.setattr("backend.runtime.manager.TradingRuntimeManager", _DummyManager)
+    runtime_state = runtime_module.get_runtime_state()
+    runtime_state.gateway_process = None
+
+    with TestClient(create_app()) as client:
+        response = client.post(
+            "/api/runtime/start",
+            json={
+                "launch_mode": "restore",
+                "restore_run_id": "20260324_100000",
+                "tickers": [],
+            },
+        )
+
+    assert response.status_code == 200
+    payload = response.json()
+    assert payload["run_id"] == "20260324_100000"
+    assert payload["run_dir"] == str(run_dir)

backend/tests/test_service_clients.py

@@ -4,6 +4,7 @@
 import pytest
 
 from shared.client.control_client import ControlPlaneClient
+from shared.client.openclaw_client import OpenClawServiceClient
 from shared.client.runtime_client import RuntimeServiceClient
 
 
@@ -105,3 +106,25 @@ async def test_runtime_service_client_hits_current_runtime_routes():
         ("get", "/config", None),
         ("put", "/config", {"schedule_mode": "intraday"}),
     ]
+
+
+@pytest.mark.asyncio
+async def test_openclaw_service_client_hits_current_openclaw_routes():
+    client = OpenClawServiceClient()
+    client._client = _DummyAsyncClient()
+
+    await client.fetch_status()
+    await client.list_sessions()
+    await client.get_session("main/session-1")
+    await client.get_session_history("main/session-1", limit=5)
+    await client.list_cron_jobs()
+    await client.list_approvals()
+
+    assert client._client.calls == [
+        ("get", "/status", None),
+        ("get", "/sessions", None),
+        ("get", "/sessions/main/session-1", None),
+        ("get", "/sessions/main/session-1/history", {"limit": 5}),
+        ("get", "/cron", None),
+        ("get", "/approvals", None),
+    ]

backend/tools/technical_signals.py

@@ -48,7 +48,7 @@ class TechnicalSignal:
 
 
 class StockTechnicalAnalyzer:
-    """Lightweight technical analyzer adapted for EvoTraders tools."""
+    """Lightweight technical analyzer adapted for 大时代 tools."""
 
     def analyze(self, ticker: str, df: pd.DataFrame) -> TechnicalSignal:
         """Analyze one ticker from OHLC price history."""

backend/utils/settlement.py

@@ -228,12 +228,12 @@ class SettlementCoordinator:
 
         all_evaluations = {**analyst_evaluations, **pm_evaluations}
 
-        leaderboard = self.storage.load_file("leaderboard") or []
+        leaderboard = self.storage.load_export_file("leaderboard") or []
         updated_leaderboard = update_leaderboard_with_evaluations(
             leaderboard,
             all_evaluations,
         )
-        self.storage.save_file("leaderboard", updated_leaderboard)
+        self.storage.save_export_file("leaderboard", updated_leaderboard)
 
         self._update_summary_with_baselines(
             date,

backend/utils/terminal_dashboard.py

@@ -1,359 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Terminal Dashboard - Persistent unified panel using Rich Live
-"""
-# pylint: disable=R0915,R0912
-import logging
-import threading
-import time
-from typing import Any, Dict, List, Optional
-
-from rich.console import Console
-from rich.live import Live
-from rich.panel import Panel
-from rich.table import Table
-
-logger = logging.getLogger(__name__)
-
-
-class TerminalDashboard:
-    """Unified persistent terminal dashboard"""
-
-    def __init__(self, console: Console = None):
-        self.console = console or Console()
-        self.live: Optional[Live] = None
-
-        # Config state
-        self.mode = "live"
-        self.config_name = ""
-        self.host = "0.0.0.0"
-        self.port = 8765
-        self.poll_interval = 10
-        self.trigger_time = "now"
-        self.mock = False
-        self.enable_memory = False
-        self.local_time = ""
-        self.nyse_time = ""
-        self.start_date = ""
-        self.end_date = ""
-        self.tickers: List[str] = []
-        self.initial_cash = 100000.0
-        self.data_sources: Dict[str, Any] = {}
-
-        # Trading state
-        self.current_date = "-"
-        self.status = "Initializing"
-        self.total_value = 0.0
-        self.cash = 0.0
-        self.pnl_pct = 0.0
-        self.holdings: List[Dict] = []
-        self.trades: List[Dict] = []
-        self.days_completed = 0
-        self.days_total = 0
-
-        # Progress message (last line)
-        self.progress = ""
-        self._dots_index = 0
-        self._animator_running = False
-        self._animator_thread: Optional[threading.Thread] = None
-
-    def set_config(
-        self,
-        mode: str,
-        config_name: str,
-        host: str,
-        port: int,
-        poll_interval: int,
-        trigger_time: str = "now",
-        mock: bool = False,
-        enable_memory: bool = False,
-        local_time: str = "",
-        nyse_time: str = "",
-        start_date: str = "",
-        end_date: str = "",
-        tickers: List[str] = None,
-        initial_cash: float = 100000.0,
-        data_sources: Dict[str, Any] = None,
-    ):
-        """Set configuration state"""
-        self.mode = mode
-        self.config_name = config_name
-        self.host = host
-        self.port = port
-        self.poll_interval = poll_interval
-        self.trigger_time = trigger_time
-        self.mock = mock
-        self.enable_memory = enable_memory
-        self.local_time = local_time
-        self.nyse_time = nyse_time
-        self.start_date = start_date
-        self.end_date = end_date
-        self.tickers = tickers or []
-        self.initial_cash = initial_cash
-        self.data_sources = data_sources or {}
-        self.total_value = initial_cash
-        self.cash = initial_cash
-
-    def _build_panel(self) -> Panel:
-        """Build the unified dashboard panel"""
-        # Main grid
-        main_table = Table.grid(padding=(0, 2))
-        main_table.add_column(width=28)
-        main_table.add_column(width=22)
-        main_table.add_column(width=22)
-
-        # Left: Config + Status
-        left = Table.grid(padding=(0, 0))
-        left.add_column()
-
-        # Mode line
-        if self.mode == "backtest":
-            mode_str = "[cyan]Backtest[/cyan]"
-        elif self.mock:
-            mode_str = "[yellow]MOCK[/yellow]"
-        else:
-            mode_str = "[green]LIVE[/green]"
-
-        left.add_row(f"[bold]Mode:[/bold] {mode_str}")
-        left.add_row(f"[dim]Config:[/dim] {self.config_name}")
-        left.add_row(f"[dim]Server:[/dim] {self.host}:{self.port}")
-        preferred_sources = self.data_sources.get("preferred", [])
-        if preferred_sources:
-            left.add_row(
-                f"[dim]Data:[/dim] {' -> '.join(preferred_sources)}",
-            )
-
-        if self.mode == "live" and self.nyse_time:
-            left.add_row(f"[dim]NYSE:[/dim] {self.nyse_time[:19]}")
-            trigger_display = (
-                "[green]NOW[/green]"
-                if self.trigger_time == "now"
-                else self.trigger_time
-            )
-            left.add_row(f"[dim]Trigger:[/dim] {trigger_display}")
-
-        # Status
-        left.add_row("")
-        status_style = "green" if self.status == "Running" else "yellow"
-        left.add_row(
-            "[bold]Status:[/bold] "
-            f"[{status_style}]{self.status}[/{status_style}]",
-        )
-        if self.mode == "backtest":
-            left.add_row(
-                f"[dim]Backtesting Period:[/dim] {self.days_total} days\n"
-                f"  {self.start_date} -> {self.end_date}",
-            )
-        left.add_row(f"[dim]Current Date:[/dim] {self.current_date}")
-
-        # Middle: Portfolio
-        mid = Table.grid(padding=(0, 0))
-        mid.add_column()
-
-        pnl_style = "green" if self.pnl_pct >= 0 else "red"
-        mid.add_row("[bold]Portfolio[/bold]")
-        mid.add_row(f"NAV: [bold]${self.total_value:,.0f}[/bold]")
-        mid.add_row(f"Cash: ${self.cash:,.0f}")
-        mid.add_row(f"P&L: [{pnl_style}]{self.pnl_pct:+.2f}%[/{pnl_style}]")
-
-        # Positions
-        mid.add_row("")
-        mid.add_row("[bold]Positions[/bold]")
-        stock_holdings = [
-            h for h in self.holdings if h.get("ticker") != "CASH"
-        ]
-        if stock_holdings:
-            for h in stock_holdings[:7]:
-                qty = h.get("quantity", 0)
-                ticker = h.get("ticker", "")[:5]
-                val = h.get("marketValue", 0)
-                qty_str = f"{qty:+d}" if qty != 0 else "0"
-                mid.add_row(
-                    f"[cyan]{ticker:<5}[/cyan] {qty_str:>5} ${val:>7,.0f}",
-                )
-            if len(stock_holdings) > 7:
-                mid.add_row(f"[dim]+{len(stock_holdings) - 7} more[/dim]")
-        else:
-            mid.add_row("[dim]No positions[/dim]")
-
-        # Right: Recent Trades
-        right = Table.grid(padding=(0, 0))
-        right.add_column()
-
-        right.add_row("[bold]Recent Trades[/bold]")
-        if self.trades:
-            for t in self.trades[:10]:
-                side = t.get("side", "")
-                ticker = t.get("ticker", "")[:5]
-                qty = t.get("qty", 0)
-                if side == "LONG":
-                    side_str = "[green]L[/green]"
-                elif side == "SHORT":
-                    side_str = "[red]S[/red]"
-                else:
-                    side_str = "[dim]H[/dim]"
-                right.add_row(f"{side_str} [cyan]{ticker:<5}[/cyan] {qty:>4}")
-            if len(self.trades) > 10:
-                right.add_row(f"[dim]+{len(self.trades) - 10} more[/dim]")
-        else:
-            right.add_row("[dim]No trades[/dim]")
-
-        main_table.add_row(left, mid, right)
-
-        # Outer table to add progress line at bottom
-        outer = Table.grid(padding=(0, 0))
-        outer.add_column()
-        outer.add_row(main_table)
-
-        # Progress line (last row) with animated dots
-        if self.progress:
-            DOTS_FRAMES = ["   ", ".  ", ".. ", "..."]
-            dots = DOTS_FRAMES[self._dots_index % len(DOTS_FRAMES)]
-            outer.add_row("")
-            outer.add_row(f"[dim]> {self.progress}{dots}[/dim]")
-
-        # Build panel
-        title = "[bold cyan]EvoTraders[/bold cyan]"
-        if self.mode == "backtest":
-            title += " [dim]Backtest[/dim]"
-        elif self.mock:
-            title += " [dim]Mock[/dim]"
-        else:
-            title += " [dim]Live[/dim]"
-
-        return Panel(
-            outer,
-            title=title,
-            border_style="cyan",
-            padding=(0, 1),
-        )
-
-    def _run_animator(self):
-        """Background thread to animate the dots"""
-        while self._animator_running:
-            time.sleep(0.3)
-            if self.progress and self.live:
-                self._dots_index += 1
-                self.live.update(self._build_panel())
-
-    def start(self):
-        """Start the live dashboard display"""
-        self.live = Live(
-            self._build_panel(),
-            console=self.console,
-            refresh_per_second=4,
-            vertical_overflow="visible",
-        )
-        self.live.start()
-
-        # Start animator thread
-        self._animator_running = True
-        self._animator_thread = threading.Thread(
-            target=self._run_animator,
-            daemon=True,
-        )
-        self._animator_thread.start()
-
-    def stop(self):
-        """Stop the live dashboard"""
-        self._animator_running = False
-        if self._animator_thread:
-            self._animator_thread.join(timeout=0.5)
-            self._animator_thread = None
-        if self.live:
-            self.live.stop()
-            self.live = None
-
-    def update(
-        self,
-        date: str = None,
-        status: str = None,
-        portfolio: Dict[str, Any] = None,
-        holdings: List[Dict] = None,
-        trades: List[Dict] = None,
-        days_completed: int = None,
-        days_total: int = None,
-        data_sources: Dict[str, Any] = None,
-    ):
-        """Update dashboard state and refresh display"""
-        if date:
-            self.current_date = date
-        if status:
-            self.status = status
-        if days_completed is not None:
-            self.days_completed = days_completed
-        if days_total is not None:
-            self.days_total = days_total
-
-        if portfolio:
-            self.total_value = portfolio.get(
-                "totalAssetValue",
-                0,
-            ) or portfolio.get(
-                "total_value",
-                self.initial_cash,
-            )
-            self.cash = portfolio.get("cashPosition", 0) or portfolio.get(
-                "cash",
-                self.initial_cash,
-            )
-            if self.total_value > 0 and self.initial_cash > 0:
-                self.pnl_pct = (
-                    (self.total_value - self.initial_cash) / self.initial_cash
-                ) * 100
-
-        if holdings is not None:
-            self.holdings = holdings
-        if trades is not None:
-            self.trades = trades
-        if data_sources is not None:
-            self.data_sources = data_sources
-
-        if self.live:
-            self.live.update(self._build_panel())
-
-    def log(self, msg: str, also_log: bool = True):
-        """
-        Update progress message and refresh panel
-
-        Args:
-            msg: Progress message to display
-            also_log: Whether to also write to logger (default True)
-        """
-        self.progress = msg
-        if also_log:
-            logger.info(msg)
-        if self.live:
-            self.live.update(self._build_panel())
-
-    def print_final_summary(self):
-        """Print final summary when dashboard stops"""
-        pnl_style = "green" if self.pnl_pct >= 0 else "red"
-
-        if self.mode == "backtest":
-            msg = (
-                f"[bold]Backtest Complete[/bold] | "
-                f"Days: {self.days_completed} | "
-                f"NAV: ${self.total_value:,.0f} | "
-                f"Return: [{pnl_style}]{self.pnl_pct:+.2f}%[/{pnl_style}]"
-            )
-        else:
-            msg = (
-                f"[bold]Session End[/bold] | "
-                f"NAV: ${self.total_value:,.0f} | "
-                f"P&L: [{pnl_style}]{self.pnl_pct:+.2f}%[/{pnl_style}]"
-            )
-
-        self.console.print(Panel(msg, border_style="green"))
-
-
-# Global instance
-_dashboard: Optional[TerminalDashboard] = None
-
-
-def get_dashboard() -> TerminalDashboard:
-    """Get or create global dashboard instance"""
-    global _dashboard
-    if _dashboard is None:
-        _dashboard = TerminalDashboard()
-    return _dashboard

backtest/state/server_state.json

@@ -2626,6 +2626,5 @@
   "trading_days_completed": 5,
   "server_mode": "backtest",
   "is_backtest": true,
-  "is_mock_mode": false,
   "last_saved": "2026-03-12T23:07:31.098122"
 }

deploy/README.md

@@ -0,0 +1,135 @@
+# Deployment Notes
+
+This directory contains the current production-oriented deployment artifacts for
+the 大时代 frontend site and the live gateway process.
+
+## Contents
+
+- [deploy/systemd/evotraders.service](./systemd/evotraders.service)
+  - systemd unit for the long-running 大时代 gateway process
+- [scripts/run_prod.sh](../scripts/run_prod.sh)
+  - production launch script used by the systemd unit
+- [deploy/nginx/bigtime.cillinn.com.conf](./nginx/bigtime.cillinn.com.conf)
+  - HTTPS nginx config with WebSocket proxying
+- [deploy/nginx/bigtime.cillinn.com.http.conf](./nginx/bigtime.cillinn.com.http.conf)
+  - plain HTTP/static-site variant
+
+## Current Production Shape
+
+The checked-in production path is intentionally minimal:
+
+- nginx serves the built frontend from `/var/www/bigtime/current`
+- public domain examples use `bigtime.cillinn.com`
+- nginx proxies `/ws` to `127.0.0.1:8765`
+- systemd runs `scripts/run_prod.sh`
+- `scripts/run_prod.sh` starts `python3 -m backend.main` in live mode on `127.0.0.1:8765`
+
+This means the checked-in production example is centered on the gateway and
+frontend, not on exposing the split FastAPI services directly.
+
+## Important Paths And Ports
+
+- frontend root: `/var/www/bigtime/current`
+- gateway bind: `127.0.0.1:8765`
+- public WebSocket path: `/ws`
+- working directory expected by systemd: `/root/code/evotraders`
+
+## systemd
+
+The current systemd unit:
+
+- uses `WorkingDirectory=/root/code/evotraders`
+- executes [scripts/run_prod.sh](../scripts/run_prod.sh)
+- restarts automatically on failure
+
+Enable and start:
+
+```bash
+sudo cp deploy/systemd/evotraders.service /etc/systemd/system/evotraders.service
+sudo systemctl daemon-reload
+sudo systemctl enable evotraders
+sudo systemctl start evotraders
+```
+
+Check status and logs:
+
+```bash
+sudo systemctl status evotraders
+journalctl -u evotraders -f
+```
+
+## nginx
+
+The HTTPS nginx config does two things:
+
+- redirects `http://bigtime.cillinn.com` to HTTPS
+- proxies `/ws` to the local gateway process with WebSocket upgrade headers
+
+Typical install flow:
+
+```bash
+sudo cp deploy/nginx/bigtime.cillinn.com.conf /etc/nginx/sites-available/bigtime.cillinn.com.conf
+sudo ln -s /etc/nginx/sites-available/bigtime.cillinn.com.conf /etc/nginx/sites-enabled/
+sudo nginx -t
+sudo systemctl reload nginx
+```
+
+The checked-in TLS config expects Let's Encrypt assets at:
+
+- `/etc/letsencrypt/live/bigtime.cillinn.com/fullchain.pem`
+- `/etc/letsencrypt/live/bigtime.cillinn.com/privkey.pem`
+
+## Environment Expectations
+
+Before using the production scripts, ensure the runtime environment has:
+
+- a usable Python environment
+- backend dependencies installed from `requirements.txt`
+- the package installed with `pip install -e .` or `uv pip install -e .`
+- frontend dependencies installed with `npm ci`
+- repo dependencies installed
+- required market/model API keys
+- any desired `TICKERS` override
+
+Recommended production install sequence:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -r requirements.txt
+pip install -e .
+cd frontend && npm ci && npm run build && cd ..
+```
+
+The production script currently sets:
+
+```bash
+PYTHONPATH=/root/code/evotraders/.pydeps:.
+TICKERS=${TICKERS:-AAPL,MSFT,GOOGL,AMZN,NVDA,META,TSLA,AMD,NFLX,AVGO,PLTR,COIN}
+```
+
+It then launches:
+
+```bash
+python3 -m backend.main \
+  --mode live \
+  --config-name production \
+  --host 127.0.0.1 \
+  --port 8765 \
+  --trigger-time now \
+  --poll-interval 15
+```
+
+## What This Deployment Does Not Yet Cover
+
+The checked-in deployment artifacts do not currently document or automate:
+
+- split FastAPI service deployment on `8000` to `8004`
+- OpenClaw gateway deployment on `18789`
+- database backup/retention workflows
+- frontend build/publish steps
+- secret management
+
+If you move production fully to split-service mode, update this directory so it
+documents the new service topology explicitly instead of relying on the gateway-
+only path.

deploy/nginx/bigtime.cillinn.com.conf

@@ -1,9 +1,9 @@
 server {
     listen 80;
-    server_name evotraders.cillinn.com;
+    server_name bigtime.cillinn.com;
 
     location /.well-known/acme-challenge/ {
-        root /var/www/evotraders/current;
+        root /var/www/bigtime/current;
         allow all;
     }
 
@@ -14,13 +14,13 @@ server {
 
 server {
     listen 443 ssl http2;
-    server_name evotraders.cillinn.com;
+    server_name bigtime.cillinn.com;
 
-    root /var/www/evotraders/current;
+    root /var/www/bigtime/current;
     index index.html;
 
-    ssl_certificate /etc/letsencrypt/live/evotraders.cillinn.com/fullchain.pem;
+    ssl_certificate /etc/letsencrypt/live/bigtime.cillinn.com/fullchain.pem;
-    ssl_certificate_key /etc/letsencrypt/live/evotraders.cillinn.com/privkey.pem;
+    ssl_certificate_key /etc/letsencrypt/live/bigtime.cillinn.com/privkey.pem;
     include /etc/letsencrypt/options-ssl-nginx.conf;
     ssl_dhparam /etc/letsencrypt/ssl-dhparams.pem;
 

deploy/nginx/bigtime.cillinn.com.http.conf

@@ -1,8 +1,8 @@
 server {
     listen 80;
-    server_name evotraders.cillinn.com;
+    server_name bigtime.cillinn.com;
 
-    root /var/www/evotraders/current;
+    root /var/www/bigtime/current;
     index index.html;
 
     location /.well-known/acme-challenge/ {

deploy/systemd/evotraders.service

@@ -1,5 +1,5 @@
 [Unit]
-Description=EvoTraders Production Service
+Description=大时代 Production Service
 After=network.target
 
 [Service]

docs/compat-removal-plan.md

@@ -1,28 +1,116 @@
-# Compatibility Removal Plan
+# Compatibility And Migration Status
 
-This document tracks the remaining migration-only surfaces that still exist
+This document tracks the remaining migration-related boundaries after the
-after the move to split-first development.
+repository switched to split-first development.
 
-## Migration-only Surfaces
+## Current Status
 
-None currently remain as dedicated compatibility wrappers.
+The repo no longer depends on a combined FastAPI compatibility wrapper for
+normal local development. The default path is now:
 
-## Completed Removals
+`agent_service + trading_service + news_service + runtime_service + gateway`
+
+That means compatibility is no longer a separate startup mode. What remains is
+mostly protocol-level and routing-level compatibility while the codebase
+continues to move responsibilities into clearer service surfaces.
+
+## What Was Removed
 
 ### `backend.app`
 
-- Removed after compatibility startup switched to
+- Removed after startup paths switched away from the legacy app wrapper.
-  `backend.apps.combined_service:app` directly.
+
+### `backend.apps.combined_service`
+
+- Removed after split-service startup became the only supported local dev mode.
 
 ### `shared.client.AgentServiceClient`
 
 - Removed after split-aware clients became the default import surface.
-- Replacement:
+- Replaced by:
   - `ControlPlaneClient`
   - `RuntimeServiceClient`
   - `TradingServiceClient`
   - `NewsServiceClient`
 
-### `backend.apps.combined_service`
+## What Still Exists For Compatibility
 
-- Removed after split-service mode became the only supported dev startup path.
+These are not legacy wrappers in the old sense, but they still preserve
+backward-compatible behavior while migration settles.
+
+### Gateway-mediated flows
+
+- The WebSocket gateway still carries a mix of:
+  - live runtime feed transport
+  - orchestration
+  - selected read flows that have not been moved to direct browser service calls
+- This is intentional for now because the frontend still depends on the gateway
+  for event streaming and some compatibility reads.
+
+### In-process fallbacks
+
+- Some read paths still support local-module fallback when split-service URLs
+  are not configured.
+- Relevant variables include:
+  - `TRADING_SERVICE_URL`
+  - `NEWS_SERVICE_URL`
+- This keeps the app resilient during migration, but it also means behavior can
+  differ depending on env configuration.
+
+### Dual OpenClaw integration surfaces
+
+- OpenClaw currently appears through two different shapes:
+  - WebSocket gateway integration on `:18789`
+  - optional REST surface at `backend.apps.openclaw_service` on `:8004`
+- These are both valid, but they are not the same surface and should not be
+  documented as interchangeable.
+
+## Remaining Migration Risks
+
+### Split service deployment is not yet the checked-in production default
+
+- The repo documents split-service local development clearly.
+- The checked-in production example still centers on `backend.main` and nginx
+  WebSocket proxying.
+- This is a topology mismatch to keep in mind when changing deploy docs or prod
+  automation.
+
+### Environment-dependent routing
+
+- The frontend and gateway can switch behavior based on configured service URLs.
+- This is helpful operationally, but it makes debugging more configuration-
+  sensitive than a fully fixed service topology.
+
+### Runtime/control-plane separation is logical, not fully operationally isolated
+
+- `runtime_service` owns lifecycle APIs.
+- `agent_service` owns control-plane APIs.
+- The gateway still hosts the live runtime orchestration path, so the split is
+  clean at the API level but not yet a completely independent service mesh.
+
+## Exit Criteria For Declaring Migration Complete
+
+Migration can be considered effectively complete when all of the following are
+true:
+
+1. Production deployment docs and scripts explicitly run the same split-service
+   topology used in development, or intentionally document a different stable
+   production topology.
+2. Critical read paths no longer require ambiguous fallback behavior to local
+   module implementations.
+3. OpenClaw integration is documented as a stable contract with clear guidance
+   on when to use the WebSocket gateway versus the REST surface.
+4. The frontend-service routing model is stable enough that direct-service and
+   gateway-mediated paths are deliberate design choices rather than migration
+   leftovers.
+
+## Practical Read Of The Current State
+
+The migration away from combined-service startup is done.
+
+What remains is not “legacy startup debt”, but:
+
+- topology clarification
+- deployment consistency
+- reduction of env-dependent fallback behavior
+- sharper documentation around gateway and OpenClaw boundaries

env.template

@@ -1,6 +1,6 @@
 # ================== General Configuration | 通用配置 ==================
 # List of stock ticker symbols to analyze (comma-separated) | 想要分析的股票代码列表（用逗号分隔）
-TICKERS=AAPL,MSFT,GOOGL,NVDA,TSLA,META,AMZN
+TICKERS=AAPL,MSFT,GOOGL,AMZN,NVDA,META,TSLA,AMD,NFLX,AVGO,PLTR,COIN
 
 # Financial Data API
 # At least FINANCIAL_DATASETS_API_KEY is required, corresponding to FIN_DATA_SOURCE=financial_datasets; It's recommended to add FINNHUB_API_KEY, corresponding to FIN_DATA_SOURCE=finnhub; FINNHUB_API_KEY is mandatory for live mode
@@ -20,6 +20,9 @@ MARKET_DB_PATH= #optional path for long-lived market_research.db | 长期市场
 OPENAI_API_KEY=
 OPENAI_BASE_URL=
 MODEL_NAME=qwen3-max-preview
+OPENCLAW_CMD=
+OPENCLAW_CWD=
+OPENCLAW_TIMEOUT_SECONDS=15
 EXPLAIN_ENRICH_USE_LLM=false
 EXPLAIN_ENRICH_MODEL_PROVIDER=
 EXPLAIN_ENRICH_MODEL_NAME=

frontend/README.md

@@ -1,31 +1,66 @@
-## QuickStart
+## Frontend Quick Start
+
 ```bash
 cd frontend
-npm install
+npm ci
 npm run dev
 ```
 
-## Optional Direct Service Calls
+Default dev URL: `http://localhost:5173`
 
-The frontend still works with the compatibility backend entrypoint by default.
+The frontend expects the 大时代 gateway WebSocket on `ws://localhost:8765` unless overridden.
-In the current test-stage setup, split services are the recommended default.
+
-Point the frontend directly at those standalone services:
+## Recommended Local Backend Stack
+
+Start the split backend services from the project root:
+
+```bash
+./start-dev.sh
+```
+
+That gives you:
+
+- control plane at `http://localhost:8000/api`
+- trading service at `http://localhost:8001`
+- news service at `http://localhost:8002`
+- runtime service at `http://localhost:8003/api/runtime`
+- gateway WebSocket at `ws://localhost:8765`
+
+## Frontend Environment Variables
+
+You can point the frontend directly at those services with:
 
 ```bash
 VITE_CONTROL_API_BASE_URL=http://localhost:8000/api
 VITE_RUNTIME_API_BASE_URL=http://localhost:8003/api/runtime
 VITE_NEWS_SERVICE_URL=http://localhost:8002
 VITE_TRADING_SERVICE_URL=http://localhost:8001
+VITE_WS_URL=ws://localhost:8765
 ```
 
-Current direct-call coverage:
+There is also a starter template at [frontend/env.template](./env.template).
 
-- runtime panel + gateway port discovery
+For production deployments, prefer:
+
+```bash
+cd frontend
+npm ci
+npm run build
+```
+
+This ensures the deployed frontend matches the checked-in `package-lock.json`.
+
+## Direct-Service Coverage
+
+Current direct-call coverage includes:
+
+- runtime panel data loading
+- gateway port/runtime discovery
 - `story`
 - `similar days`
 - `range explain`
 - `news for date`
 - `news categories`
+- selected trading reads such as price history and insider trades
 
-If these variables are not set, the frontend falls back to the existing
+If these variables are not set, the frontend falls back to local defaults and compatibility paths where they still exist.
-WebSocket-driven compatibility flow.

frontend/env.template

@@ -1,10 +1,24 @@
 # Frontend Environment Variables Template
 # 复制此文件为 .env 并修改配置
 
-# WebSocket服务器地址
+# 控制面 API（agent/workspaces/guard）
-# 本地开发
+VITE_CONTROL_API_BASE_URL=http://localhost:8000/api
+
+# 运行时 API（start/stop/runtime info）
+VITE_RUNTIME_API_BASE_URL=http://localhost:8003/api/runtime
+
+# 新闻服务（可选，未配置时走默认回退）
+VITE_NEWS_SERVICE_URL=http://localhost:8002
+
+# 交易数据服务（可选，未配置时走默认回退）
+VITE_TRADING_SERVICE_URL=http://localhost:8001
+
+# WebSocket Gateway
 VITE_WS_URL=ws://localhost:8765
 
-# 生产环境（替换为你的实际服务器地址）
+# 生产环境示例
-# VITE_WS_URL=wss://your-server.com:8765
+# VITE_CONTROL_API_BASE_URL=https://your-domain.com/api
-
+# VITE_RUNTIME_API_BASE_URL=https://your-domain.com/api/runtime
+# VITE_NEWS_SERVICE_URL=https://your-domain.com/news
+# VITE_TRADING_SERVICE_URL=https://your-domain.com/trading
+# VITE_WS_URL=wss://your-domain.com/ws

frontend/index.html

@@ -4,7 +4,7 @@
     <meta charset="UTF-8" />
       <link rel="icon" type="image/png" href="/trading_logo.png" />
       <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>EvoTraders</title>
+    <title>大时代</title>
   </head>
   <body>
     <div id="root"></div>

frontend/package.json

@@ -3,6 +3,10 @@
   "private": true,
   "version": "0.0.0",
   "type": "module",
+  "engines": {
+    "node": ">=20",
+    "npm": ">=10"
+  },
   "scripts": {
     "dev": "vite",
     "build": "vite build",
@@ -13,6 +17,10 @@
     "preview:host": "vite preview --host"
   },
   "dependencies": {
+    "@dicebear/collection": "^9.4.2",
+    "@dicebear/core": "^9.4.2",
+    "@lobehub/icons": "^5.2.0",
+    "@lobehub/ui": "^5.6.2",
     "@radix-ui/react-dialog": "^1.1.15",
     "@radix-ui/react-dropdown-menu": "^2.1.16",
     "@radix-ui/react-label": "^2.1.7",
@@ -24,6 +32,7 @@
     "@react-three/drei": "^10.7.6",
     "@react-three/fiber": "^9.3.0",
     "@tailwindcss/vite": "^4.1.13",
+    "antd": "^6.3.4",
     "class-variance-authority": "^0.7.1",
     "clsx": "^2.1.1",
     "framer-motion": "^12.23.13",
@@ -31,6 +40,7 @@
     "lucide-react": "^0.544.0",
     "react": "^19.1.1",
     "react-dom": "^19.1.1",
+    "react-is": "^19.2.0",
     "react-markdown": "^10.1.0",
     "recharts": "^3.2.1",
     "remark-gfm": "^4.0.1",

frontend/src/components/AgentCard.jsx

@@ -1,6 +1,7 @@
 import React from 'react';
 import { ASSETS } from '../config/constants';
 import { getModelIcon, getShortModelName } from '../utils/modelIcons';
+import LobeModelLogo from './LobeModelLogo.jsx';
 
 /**
  * Get rank medal/trophy
@@ -57,7 +58,7 @@ export default function AgentCard({ agent, onClose, isClosing }) {
       background: '#ffffff',
       borderBottom: '2px solid #000000',
       boxShadow: '0 4px 12px rgba(0, 0, 0, 0.1)',
-      zIndex: 1000,
+      zIndex: 800,
       animation: isClosing ? 'slideUp 0.2s ease-out forwards' : 'slideDown 0.25s ease-out'
     }}>
       {/* Horizontal scrollable content */}
@@ -207,14 +208,18 @@ export default function AgentCard({ agent, onClose, isClosing }) {
                 justifyContent: 'center',
                 marginBottom: 4
               }}>
-                {modelInfo.logoPath ? (
+                {agent.modelName || modelInfo.logoPath ? (
-                  <img
+                  <LobeModelLogo
-                    src={modelInfo.logoPath}
+                    model={agent.modelName}
+                    provider={agent.modelProvider}
+                    fallbackSrc={modelInfo.logoPath}
                     alt={modelInfo.provider}
+                    size={36}
+                    type="color"
+                    shape="square"
                     style={{
                       maxHeight: '100%',
                       maxWidth: '100%',
-                      objectFit: 'contain'
                     }}
                   />
                 ) : (

frontend/src/components/AgentFeed.jsx

@@ -3,6 +3,7 @@ import { formatTime } from '../utils/formatters';
 import { MESSAGE_COLORS, getAgentColors, AGENTS, ASSETS } from '../config/constants';
 import { getModelIcon } from '../utils/modelIcons';
 import MarkdownModal from './MarkdownModal';
+import LobeModelLogo from './LobeModelLogo.jsx';
 
 const isAnalyst = (agentId, agentName) => {
   if (agentId && agentId.includes('analyst')) return true;
@@ -35,14 +36,22 @@ const stripMarkdown = (text) => {
     .replace(/^[-=]+$/gm, '');
 };
 
-const AgentFeed = forwardRef(({ feed, leaderboard }, ref) => {
+const AgentFeed = forwardRef(({ feed, leaderboard, agentProfilesByAgent }, ref) => {
   const feedContentRef = useRef(null);
   const [highlightedId, setHighlightedId] = useState(null);
   const [selectedAgent, setSelectedAgent] = useState('all');
   const [dropdownOpen, setDropdownOpen] = useState(false);
 
   const getAgentModelInfo = (agentId) => {
-    if (!leaderboard || !agentId) return { modelName: null, modelProvider: null };
+    if (!agentId) return { modelName: null, modelProvider: null };
+    const profile = agentProfilesByAgent?.[agentId];
+    if (profile?.model_name) {
+      return {
+        modelName: profile.model_name,
+        modelProvider: profile.model_provider
+      };
+    }
+    if (!leaderboard) return { modelName: null, modelProvider: null };
     const agentData = leaderboard.find(lb => lb.id === agentId || lb.agentId === agentId);
     return {
       modelName: agentData?.modelName,
@@ -52,7 +61,17 @@ const AgentFeed = forwardRef(({ feed, leaderboard }, ref) => {
 
   // Get agent info by name
   const getAgentInfoByName = (agentName) => {
-    if (!leaderboard || !agentName) return null;
+    if (!agentName) return null;
+    const agentConfig = AGENTS.find((agent) => agent.name === agentName);
+    const profile = agentConfig ? agentProfilesByAgent?.[agentConfig.id] : null;
+    if (agentConfig && profile?.model_name) {
+      return {
+        agentId: agentConfig.id,
+        modelName: profile.model_name,
+        modelProvider: profile.model_provider
+      };
+    }
+    if (!leaderboard) return null;
     const agentData = leaderboard.find(lb => lb.name === agentName || lb.agentName === agentName);
     if (!agentData) return null;
     return {
@@ -149,11 +168,11 @@ const AgentFeed = forwardRef(({ feed, leaderboard }, ref) => {
   // Get current selection display info
   const getCurrentSelectionInfo = () => {
     if (selectedAgent === 'all') {
-      return { label: '全部角色', modelInfo: null };
+      return { label: '全部角色', modelInfo: null, agentInfo: null };
     }
     const agentInfo = getAgentInfoByName(selectedAgent);
     const modelInfo = agentInfo ? getModelIcon(agentInfo.modelName, agentInfo.modelProvider) : null;
-    return { label: selectedAgent, modelInfo };
+    return { label: selectedAgent, modelInfo, agentInfo };
   };
 
   const currentSelection = getCurrentSelectionInfo();
@@ -171,11 +190,16 @@ const AgentFeed = forwardRef(({ feed, leaderboard }, ref) => {
               onBlur={() => setTimeout(() => setDropdownOpen(false), 200)}
             >
               <div className="custom-select-value">
-                {currentSelection.modelInfo?.logoPath && (
+                {(currentSelection.agentInfo?.modelName || currentSelection.modelInfo?.logoPath) && (
-                  <img
+                  <LobeModelLogo
-                    src={currentSelection.modelInfo.logoPath}
+                    model={currentSelection.agentInfo?.modelName}
-                    alt={currentSelection.modelInfo.provider}
+                    provider={currentSelection.agentInfo?.modelProvider}
+                    fallbackSrc={currentSelection.modelInfo?.logoPath}
+                    alt={currentSelection.modelInfo?.provider}
+                    size={18}
                     className="select-model-icon"
+                    shape="square"
+                    type="color"
                   />
                 )}
                 <span>{currentSelection.label}</span>
@@ -205,11 +229,16 @@ const AgentFeed = forwardRef(({ feed, leaderboard }, ref) => {
                         setDropdownOpen(false);
                       }}
                     >
-                      {modelInfo?.logoPath && (
+                      {(agentInfo?.modelName || modelInfo?.logoPath) && (
-                        <img
+                        <LobeModelLogo
-                          src={modelInfo.logoPath}
+                          model={agentInfo?.modelName}
-                          alt={modelInfo.provider}
+                          provider={agentInfo?.modelProvider}
+                          fallbackSrc={modelInfo?.logoPath}
+                          alt={modelInfo?.provider}
+                          size={18}
                           className="select-model-icon"
+                          shape="square"
+                          type="color"
                         />
                       )}
                       <span>{agent}</span>
@@ -345,16 +374,16 @@ function ConferenceMessage({ message, getAgentModelInfo }) {
   return (
     <div className="conf-message-item">
       <div className="conf-agent-name" style={{ color: agentColors.text, display: 'flex', alignItems: 'center', gap: '6px', fontSize: '12px' }}>
-        {modelInfo.logoPath && (
+        {(agentModelData.modelName || modelInfo.logoPath) && (
-          <img
+          <LobeModelLogo
-            src={modelInfo.logoPath}
+            model={agentModelData.modelName}
+            provider={agentModelData.modelProvider}
+            fallbackSrc={modelInfo.logoPath}
             alt={modelInfo.provider}
-            style={{
+            size={20}
-              width: '20px',
+            shape="circle"
-              height: '20px',
+            type="color"
-              borderRadius: '50%',
+            style={{ borderRadius: '50%' }}
-              objectFit: 'contain'
-            }}
           />
         )}
         {message.agent}
@@ -573,16 +602,16 @@ function MessageItem({ message, itemId, isHighlighted, getAgentModelInfo }) {
       >
         <div className="feed-item-header">
         <span className="feed-item-title" style={{ color: colors.text, display: 'flex', alignItems: 'center', gap: '6px', fontSize: '12px' }}>
-            {modelInfo.logoPath && message.agent !== 'Memory' && (
+            {message.agent !== 'Memory' && (agentModelData.modelName || modelInfo.logoPath) && (
-              <img
+              <LobeModelLogo
-                src={modelInfo.logoPath}
+                model={agentModelData.modelName}
+                provider={agentModelData.modelProvider}
+                fallbackSrc={modelInfo.logoPath}
                 alt={modelInfo.provider}
-                style={{
+                size={20}
-                  width: '20px',
+                shape="circle"
-                  height: '20px',
+                type="color"
-                  borderRadius: '50%',
+                style={{ borderRadius: '50%' }}
-                  objectFit: 'contain'
-                }}
               />
             )}
             {title}

frontend/src/components/AppShell.jsx

@@ -0,0 +1,511 @@
+import React, { Suspense, lazy, useRef, useEffect, useMemo } from 'react';
+import GlobalStyles from '../styles/GlobalStyles';
+import Header from './Header.jsx';
+import RuntimeSettingsPanel from './RuntimeSettingsPanel.jsx';
+import NetValueChart from './NetValueChart.jsx';
+import { AGENTS } from '../config/constants';
+import { useRuntimeStore } from '../store/runtimeStore';
+import { useUIStore } from '../store/uiStore';
+import { formatNumber, formatTickerPrice } from '../utils/formatters';
+
+const RoomView = lazy(() => import('./RoomView'));
+const AgentFeed = lazy(() => import('./AgentFeed'));
+const StatisticsView = lazy(() => import('./StatisticsView'));
+const StockExplainView = lazy(() => import('./StockExplainView.jsx'));
+const TraderView = lazy(() => import('./TraderView.jsx'));
+
+function ViewLoadingFallback({ label = '加载中...' }) {
+  return (
+    <div style={{
+      minHeight: 240,
+      height: '100%',
+      display: 'flex',
+      alignItems: 'center',
+      justifyContent: 'center',
+      border: '1px solid #000000',
+      background: '#ffffff',
+      fontSize: 12,
+      fontWeight: 700,
+      letterSpacing: 0.4
+    }}>
+      {label}
+    </div>
+  );
+}
+
+/**
+ * AppShell - Layout shell containing Header, TickerBar, ViewNavBar, View container, and AgentFeed
+ */
+export default function AppShell({
+  // Connection & status
+  isConnected,
+  virtualTime,
+  now,
+  marketStatus,
+  serverMode,
+  marketStatusLabel,
+  dataSourceLabel,
+  runtimeSummaryLabel,
+  isUpdating,
+  // Handlers
+  onManualTrigger,
+  onOpenRuntimeLogs,
+  onRuntimeSettingsToggle,
+  // Runtime settings panel props
+  isRuntimeSettingsOpen,
+  isRuntimeConfigSaving,
+  isWatchlistSaving,
+  runtimeConfigFeedback,
+  watchlistFeedback,
+  launchModeDraft,
+  restoreRunIdDraft,
+  runtimeHistoryRuns,
+  scheduleModeDraft,
+  intervalMinutesDraft,
+  triggerTimeDraft,
+  maxCommCyclesDraft,
+  initialCashDraft,
+  marginRequirementDraft,
+  enableMemoryDraft,
+  modeDraft,
+  pollIntervalDraft,
+  startDateDraft,
+  endDateDraft,
+  watchlistDraftSymbols,
+  watchlistInputValue,
+  watchlistSuggestions,
+  onLaunchModeChange,
+  onRestoreRunIdChange,
+  onScheduleModeChange,
+  onIntervalMinutesChange,
+  onTriggerTimeChange,
+  onMaxCommCyclesChange,
+  onInitialCashChange,
+  onMarginRequirementChange,
+  onEnableMemoryChange,
+  onModeChange,
+  onPollIntervalChange,
+  onStartDateChange,
+  onEndDateChange,
+  onWatchlistInputChange,
+  onWatchlistInputKeyDown,
+  onWatchlistAdd,
+  onWatchlistRemove,
+  onWatchlistRestoreCurrent,
+  onWatchlistRestoreDefault,
+  onWatchlistSuggestionClick,
+  onLaunchConfigSave,
+  onRestoreDefaults,
+  // Ticker and portfolio data
+  displayTickers,
+  portfolioData,
+  rollingTickers,
+  // Feed data
+  feed,
+  bubbles,
+  bubbleFor,
+  leaderboard,
+  // Views data
+  currentView,
+  chartTab,
+  holdings,
+  trades,
+  stats,
+  priceHistoryByTicker,
+  ohlcHistoryByTicker,
+  selectedExplainSymbol,
+  onSelectedExplainSymbolChange,
+  historySourceByTicker,
+  explainEventsByTicker,
+  newsByTicker,
+  insiderTradesByTicker,
+  technicalIndicatorsByTicker,
+  currentDate,
+  // Stock request handlers
+  stockRequests,
+  // Agent request handlers
+  agentRequests,
+  agentProfilesByAgent,
+  // Layout
+  leftWidth,
+  isResizing,
+  onMouseDown,
+  agentFeedRef
+}) {
+  const containerRef = useRef(null);
+  const { setIsRuntimeSettingsOpen, setIsWatchlistPanelOpen } = useRuntimeStore();
+  const { setChartTab, setCurrentView, setIsResizing, setLeftWidth } = useUIStore();
+
+  useEffect(() => {
+    if (currentView === 'openclaw') {
+      setCurrentView('statistics');
+    }
+  }, [currentView, setCurrentView]);
+
+  // Resize handler
+  useEffect(() => {
+    if (!isResizing) return;
+
+    const handleMouseMove = (e) => {
+      if (!containerRef.current) return;
+      const containerRect = containerRef.current.getBoundingClientRect();
+      const newLeftWidth = ((e.clientX - containerRect.left) / containerRect.width) * 100;
+      if (newLeftWidth >= 30 && newLeftWidth <= 85) {
+        setLeftWidth(newLeftWidth);
+      }
+    };
+
+    const handleMouseUp = () => setIsResizing(false);
+
+    document.addEventListener('mousemove', handleMouseMove);
+    document.addEventListener('mouseup', handleMouseUp);
+
+    return () => {
+      document.removeEventListener('mousemove', handleMouseMove);
+      document.removeEventListener('mouseup', handleMouseUp);
+    };
+  }, [isResizing, setIsResizing, setLeftWidth]);
+
+  const handleJumpToMessage = (bubble) => {
+    if (agentFeedRef.current && agentFeedRef.current.scrollToMessage) {
+      agentFeedRef.current.scrollToMessage(bubble);
+    }
+  };
+
+  const viewClassName = useMemo(() => {
+    const base = `view-slider-five ${currentView === 'traders' ? 'show-traders' :
+      currentView === 'room' ? 'show-room' :
+        currentView === 'explain' ? 'show-explain' :
+          currentView === 'chart' ? 'show-chart' :
+            'show-statistics'}`;
+    return base;
+  }, [currentView]);
+
+  return (
+    <div className="app">
+      <GlobalStyles />
+
+      {/* Header */}
+      <div className="header">
+        <Header />
+
+        <div className="header-right" style={{ display: 'flex', alignItems: 'center', gap: 24, marginLeft: 'auto', flexWrap: 'wrap', minWidth: 0 }}>
+          {/* Unified Status Indicator */}
+          <div className="header-status-inline">
+            <span className={`status-dot ${isConnected ? (isUpdating ? 'updating' : 'live') : 'offline'}`} />
+            <span className={`status-text ${isConnected ? 'live' : 'offline'}`}>
+              {isConnected ? (isUpdating ? '同步中' : '在线') : '离线'}
+            </span>
+            {marketStatus && (
+              <>
+                <span className="status-sep">·</span>
+                <span className={`market-text ${serverMode === 'backtest' ? 'backtest' : (marketStatus.status === 'open' ? 'open' : 'closed')}`}>
+                  {marketStatusLabel}
+                </span>
+              </>
+            )}
+            {dataSourceLabel && (
+              <>
+                <span className="status-sep">·</span>
+                <span className="market-text backtest">{dataSourceLabel}</span>
+              </>
+            )}
+            {runtimeSummaryLabel && (
+              <>
+                <span className="status-sep">·</span>
+                <span className="market-text backtest" title="当前运行配置">{runtimeSummaryLabel}</span>
+              </>
+            )}
+            <span className="status-sep">·</span>
+            <span className="time-text">{now.toLocaleTimeString('en-US', { hour: '2-digit', minute: '2-digit', second: '2-digit', hour12: false })}</span>
+          </div>
+
+          {serverMode !== 'backtest' && (
+            <div style={{ display: 'flex', gap: 8, alignItems: 'center' }}>
+              {onOpenRuntimeLogs && (
+                <button
+                  onClick={onOpenRuntimeLogs}
+                  style={{
+                    padding: '6px 12px',
+                    borderRadius: 4,
+                    background: '#FFFFFF',
+                    border: '1px solid #111111',
+                    color: '#111111',
+                    fontSize: '11px',
+                    fontFamily: '"Courier New", monospace',
+                    fontWeight: 700,
+                    cursor: 'pointer',
+                    letterSpacing: '0.4px',
+                    textTransform: 'uppercase'
+                  }}
+                  title="查看当前运行日志"
+                >
+                  运行日志
+                </button>
+              )}
+              <button
+                onClick={onManualTrigger}
+                disabled={!isConnected}
+                style={{
+                  padding: '6px 12px',
+                  borderRadius: 4,
+                  background: isConnected ? '#111111' : '#8a8a8a',
+                  border: '1px solid #111111',
+                  color: '#FFFFFF',
+                  fontSize: '11px',
+                  fontFamily: '"Courier New", monospace',
+                  fontWeight: 700,
+                  cursor: isConnected ? 'pointer' : 'not-allowed',
+                  letterSpacing: '0.4px',
+                  textTransform: 'uppercase'
+                }}
+                title="手动触发一轮分析与交易决策"
+              >
+                手动运行
+              </button>
+            </div>
+          )}
+
+          <RuntimeSettingsPanel
+            showTrigger={false}
+            isOpen={isRuntimeSettingsOpen}
+            isConnected={isConnected}
+            isSaving={isRuntimeConfigSaving || isWatchlistSaving}
+            feedback={runtimeConfigFeedback || watchlistFeedback}
+            launchMode={launchModeDraft}
+            restoreRunId={restoreRunIdDraft}
+            runtimeHistoryRuns={runtimeHistoryRuns}
+            scheduleMode={scheduleModeDraft}
+            intervalMinutes={intervalMinutesDraft}
+            triggerTime={triggerTimeDraft}
+            maxCommCycles={maxCommCyclesDraft}
+            initialCash={initialCashDraft}
+            marginRequirement={marginRequirementDraft}
+            enableMemory={enableMemoryDraft}
+            mode={modeDraft}
+            pollInterval={pollIntervalDraft}
+            startDate={startDateDraft}
+            endDate={endDateDraft}
+            watchlistSymbols={watchlistDraftSymbols}
+            watchlistInputValue={watchlistInputValue}
+            watchlistSuggestions={watchlistSuggestions}
+            onToggle={onRuntimeSettingsToggle}
+            onClose={() => setIsRuntimeSettingsOpen(false)}
+            onLaunchModeChange={onLaunchModeChange}
+            onRestoreRunIdChange={onRestoreRunIdChange}
+            onScheduleModeChange={onScheduleModeChange}
+            onIntervalMinutesChange={onIntervalMinutesChange}
+            onTriggerTimeChange={onTriggerTimeChange}
+            onMaxCommCyclesChange={onMaxCommCyclesChange}
+            onInitialCashChange={onInitialCashChange}
+            onMarginRequirementChange={onMarginRequirementChange}
+            onEnableMemoryChange={onEnableMemoryChange}
+            onModeChange={onModeChange}
+            onPollIntervalChange={onPollIntervalChange}
+            onStartDateChange={onStartDateChange}
+            onEndDateChange={onEndDateChange}
+            onWatchlistInputChange={onWatchlistInputChange}
+            onWatchlistInputKeyDown={onWatchlistInputKeyDown}
+            onWatchlistAdd={onWatchlistAdd}
+            onWatchlistRemove={onWatchlistRemove}
+            onWatchlistRestoreCurrent={onWatchlistRestoreCurrent}
+            onWatchlistRestoreDefault={onWatchlistRestoreDefault}
+            onWatchlistSuggestionClick={onWatchlistSuggestionClick}
+            onSave={onLaunchConfigSave}
+            onRestoreDefaults={onRestoreDefaults}
+          />
+        </div>
+      </div>
+
+      {/* Main Content */}
+      <>
+        {/* Ticker Bar */}
+        <div className="ticker-bar">
+          <div className="ticker-track">
+            {[0, 1].map((groupIdx) => (
+              <div key={groupIdx} className="ticker-group">
+                {displayTickers.map(ticker => (
+                  <div key={`${ticker.symbol}-${groupIdx}`} className="ticker-item">
+                    <span className="ticker-symbol">{ticker.symbol}</span>
+                    <span className="ticker-price">
+                      <span className={`ticker-price-value ${rollingTickers[ticker.symbol] ? 'rolling' : ''}`}>
+                        {ticker.price !== null && ticker.price !== undefined
+                          ? `$${formatTickerPrice(ticker.price)}` : '-'}
+                      </span>
+                    </span>
+                    <span className={`ticker-change ${
+                      ticker.change === null || ticker.change === undefined
+                        ? '' : ticker.change >= 0 ? 'positive' : 'negative'
+                    }`}>
+                      {ticker.change !== null && ticker.change !== undefined
+                        ? `${ticker.change >= 0 ? '+' : ''}${ticker.change.toFixed(2)}%` : '-'}
+                    </span>
+                  </div>
+                ))}
+              </div>
+            ))}
+          </div>
+          <div className="portfolio-value">
+            <span className="portfolio-label">投资组合</span>
+            <span className="portfolio-amount">${formatNumber(portfolioData.netValue)}</span>
+          </div>
+        </div>
+
+        <div className="main-container" ref={containerRef}>
+          {/* Left Panel */}
+          <div className="left-panel" style={{ width: `${leftWidth}%` }}>
+            <div className="chart-section">
+              <div className="view-container">
+                <div className="view-nav-bar">
+                  <button
+                    className={`view-nav-btn ${currentView === 'traders' ? 'active' : ''}`}
+                    onClick={() => setCurrentView('traders')}
+                  >
+                    交易员
+                  </button>
+                  <button
+                    className={`view-nav-btn ${currentView === 'room' ? 'active' : ''}`}
+                    onClick={() => setCurrentView('room')}
+                  >
+                    交易室
+                  </button>
+                  <button
+                    className={`view-nav-btn ${currentView === 'explain' ? 'active' : ''}`}
+                    onClick={() => setCurrentView('explain')}
+                  >
+                    个股分析
+                  </button>
+                  <button
+                    className={`view-nav-btn ${currentView === 'chart' ? 'active' : ''}`}
+                    onClick={() => setCurrentView('chart')}
+                  >
+                    业绩图表
+                  </button>
+                  <button
+                    className={`view-nav-btn ${currentView === 'statistics' ? 'active' : ''}`}
+                    onClick={() => setCurrentView('statistics')}
+                  >
+                    统计
+                  </button>
+                </div>
+
+                <div className={viewClassName}>
+                  {/* Traders View */}
+                  <div className="view-panel">
+                    <Suspense fallback={<ViewLoadingFallback label="加载交易员视图..." />}>
+                      <TraderView {...agentRequests} />
+                    </Suspense>
+                  </div>
+
+                  {/* Room View Panel */}
+                  <div className="view-panel">
+                    <Suspense fallback={<ViewLoadingFallback label="加载交易室..." />}>
+                      <RoomView
+                        bubbles={bubbles}
+                        bubbleFor={bubbleFor}
+                        leaderboard={leaderboard}
+                        agentProfilesByAgent={agentProfilesByAgent}
+                        feed={feed}
+                        onJumpToMessage={handleJumpToMessage}
+                        onOpenLaunchConfig={() => setIsRuntimeSettingsOpen(true)}
+                      />
+                    </Suspense>
+                  </div>
+
+                  {/* Stock Explain View Panel */}
+                  <div className="view-panel">
+                    <Suspense fallback={<ViewLoadingFallback label="加载个股分析..." />}>
+                      <StockExplainView
+                        tickers={displayTickers}
+                        holdings={holdings}
+                        trades={trades}
+                        leaderboard={leaderboard}
+                        feed={feed}
+                        priceHistoryByTicker={priceHistoryByTicker}
+                        ohlcHistoryByTicker={ohlcHistoryByTicker}
+                        selectedSymbol={selectedExplainSymbol}
+                        onSelectedSymbolChange={onSelectedExplainSymbolChange}
+                        selectedHistorySource={historySourceByTicker[selectedExplainSymbol] || null}
+                        explainEventsSnapshot={explainEventsByTicker[selectedExplainSymbol] || null}
+                        newsSnapshot={newsByTicker[selectedExplainSymbol] || null}
+                        insiderTradesSnapshot={insiderTradesByTicker[selectedExplainSymbol] || null}
+                        technicalIndicatorsSnapshot={technicalIndicatorsByTicker[selectedExplainSymbol] || null}
+                        onRequestHistory={stockRequests?.requestStockHistory}
+                        onRequestExplainEvents={stockRequests?.requestStockExplainEvents}
+                        onRequestNews={stockRequests?.requestStockNews}
+                        onRequestRangeExplain={stockRequests?.requestStockRangeExplain}
+                        onRequestNewsForDate={stockRequests?.requestStockNewsForDate}
+                        onRequestStory={stockRequests?.requestStockStory}
+                        onRequestInsiderTrades={stockRequests?.requestStockInsiderTrades}
+                        onRequestTechnicalIndicators={stockRequests?.requestStockTechnicalIndicators}
+                        currentDate={currentDate}
+                        onRequestSimilarDays={stockRequests?.requestStockSimilarDays}
+                        onRequestStockEnrich={stockRequests?.requestStockEnrich}
+                      />
+                    </Suspense>
+                  </div>
+
+                  {/* Chart View Panel */}
+                  <div className="view-panel">
+                    <div className="chart-container">
+                      <div className="chart-tabs-floating">
+                        <button
+                          className={`chart-tab ${chartTab === 'all' ? 'active' : ''}`}
+                          onClick={() => setChartTab('all')}
+                        >
+                          日线
+                        </button>
+                      </div>
+                      {currentView === 'chart' ? (
+                        <NetValueChart
+                          equity={portfolioData.equity}
+                          baseline={portfolioData.baseline}
+                          baseline_vw={portfolioData.baseline_vw}
+                          momentum={portfolioData.momentum}
+                          strategies={portfolioData.strategies}
+                          equity_return={portfolioData.equity_return}
+                          baseline_return={portfolioData.baseline_return}
+                          baseline_vw_return={portfolioData.baseline_vw_return}
+                          momentum_return={portfolioData.momentum_return}
+                          chartTab={chartTab}
+                          virtualTime={virtualTime}
+                        />
+                      ) : (
+                        <div style={{ height: '100%', minHeight: 320 }} />
+                      )}
+                    </div>
+                  </div>
+
+                  {/* Statistics View Panel */}
+                  <div className="view-panel">
+                    <Suspense fallback={<ViewLoadingFallback label="加载统计视图..." />}>
+                      <StatisticsView
+                        trades={trades}
+                        holdings={holdings}
+                        stats={stats}
+                        portfolioData={portfolioData}
+                        baseline_vw={portfolioData.baseline_vw}
+                        equity={portfolioData.equity}
+                        leaderboard={leaderboard}
+                      />
+                    </Suspense>
+                  </div>
+                </div>
+              </div>
+            </div>
+          </div>
+
+          {/* Resizer */}
+          <div className={`resizer ${isResizing ? 'resizing' : ''}`} onMouseDown={onMouseDown} />
+
+          {/* Right Panel: Agent Feed */}
+          <div className="right-panel" style={{ width: `${100 - leftWidth}%` }}>
+            <Suspense fallback={<ViewLoadingFallback label="加载消息流..." />}>
+              <AgentFeed ref={agentFeedRef} feed={feed} leaderboard={leaderboard} agentProfilesByAgent={agentProfilesByAgent} />
+            </Suspense>
+          </div>
+        </div>
+      </>
+    </div>
+  );
+}

frontend/src/components/Header.jsx

@@ -2,7 +2,7 @@ import React from 'react';
 
 /**
  * Header Component
- * Reusable header brand for EvoTraders.
+ * Reusable header brand for 大时代.
  */
 export default function Header() {
   return (
@@ -19,10 +19,10 @@ export default function Header() {
       >
         <img
           src="/trading_logo.png"
-          alt="EvoTraders Logo"
+          alt="大时代 Logo"
           style={{ height: '24px', width: 'auto' }}
         />
-        EvoTraders
+        大时代
       </span>
     </div>
   );

frontend/src/components/LobeModelLogo.jsx

@@ -0,0 +1,78 @@
+import React from 'react';
+import ModelIcon from '@lobehub/icons/es/features/ModelIcon';
+import ProviderIcon from '@lobehub/icons/es/features/ProviderIcon';
+
+export default function LobeModelLogo({
+  model,
+  provider,
+  fallbackSrc = null,
+  alt = '',
+  size = 28,
+  shape = 'square',
+  type = 'color',
+  style = {},
+  className = '',
+}) {
+  const hasModel = typeof model === 'string' && model.trim().length > 0;
+  const hasProvider = typeof provider === 'string' && provider.trim().length > 0;
+
+  try {
+    if (hasModel) {
+      return (
+        <ModelIcon
+          model={model}
+          size={size}
+          shape={shape}
+          type={type}
+          className={className}
+          style={style}
+        />
+      );
+    }
+
+    if (hasProvider) {
+      return (
+        <ProviderIcon
+          provider={provider.toLowerCase()}
+          size={size}
+          shape={shape}
+          type={type}
+          className={className}
+          style={style}
+        />
+      );
+    }
+  } catch {
+    // Fall through to local fallback asset.
+  }
+
+  if (fallbackSrc) {
+    return (
+      <img
+        src={fallbackSrc}
+        alt={alt}
+        className={className}
+        style={{
+          width: size,
+          height: size,
+          objectFit: 'contain',
+          ...style,
+        }}
+      />
+    );
+  }
+
+  return (
+    <div
+      className={className}
+      style={{
+        width: size,
+        height: size,
+        borderRadius: shape === 'circle' ? '50%' : 8,
+        background: '#F3F4F6',
+        border: '1px solid #D1D5DB',
+        ...style,
+      }}
+    />
+  );
+}

frontend/src/components/NetValueChart.jsx

@@ -5,10 +5,10 @@ import { formatNumber, formatFullNumber } from '../utils/formatters';
 /**
  * Helper function to get the start time of the most recent trading session
  * Trading session: 22:30 - next day 05:00
- * @param {Date|null} virtualTime - Virtual time from server (for mock mode), or null to use real time
+ * @param {Date|null} virtualTime - Virtual time from server, or null to use real time
  */
 function getRecentTradingSessionStart(virtualTime = null) {
-  // Use virtual time if provided (for mock mode), otherwise use real time
+  // Use virtual time if provided, otherwise use real time
   let now;
   if (virtualTime) {
     // Ensure virtualTime is a valid Date object
@@ -123,7 +123,7 @@ export default function NetValueChart({ equity, baseline, baseline_vw, momentum,
 
   // Legend descriptions
   const legendDescriptions = {
-    'EvoTraders': 'EvoTraders is our agents investment strategy',
+    '大时代': '大时代 is our agents investment strategy',
     'Buy & Hold (EW)': 'Equal Weight: Can be viewed as an equal-weighted index of all invested stocks',
     'Buy & Hold (VW)': 'Value Weighted: Can be viewed as a market-cap weighted index of all invested stocks',
     'Momentum': 'Momentum Strategy: Buy stocks that have performed well in the past',
@@ -758,7 +758,7 @@ export default function NetValueChart({ equity, baseline, baseline_vw, momentum,
         <Line
           type="linear"
           dataKey="portfolio"
-          name="EvoTraders"
+          name="大时代"
           stroke="#00C853"
           strokeWidth={2.5}
           dot={(props) => <CustomDot {...props} dataKey="portfolio" />}