feat: Enhance configuration management with new LLM provider support and API integration

- Backend: Introduced new endpoints for LLM configuration retrieval and updates in `config.py`, allowing dynamic management of LLM provider settings.
- Updated schemas to include `AlphaEngineConfig` for better integration with the new provider.
- Frontend: Added state management for AlphaEngine API credentials in the configuration page, ensuring seamless user experience.
- Configuration files updated to reflect changes in LLM provider settings and API keys.

BREAKING CHANGE: The default LLM provider has been changed from `new_api` to `alpha_engine`, requiring updates to existing configurations.

backend/app/routers/config.py

@@ -1,7 +1,6 @@
-"""
-API router for configuration management
-"""
 from fastapi import APIRouter, Depends, HTTPException
+from typing import Dict, Any, Optional
+from pydantic import BaseModel
 
 from app.core.dependencies import get_config_manager
 from app.schemas.config import ConfigResponse, ConfigUpdateRequest, ConfigTestRequest, ConfigTestResponse
@@ -9,11 +8,112 @@ from app.services.config_manager import ConfigManager
 
 router = APIRouter()
 
+class LLMConfigUpdate(BaseModel):
+    provider: str
+    model: Optional[str] = None
+
 @router.get("/", response_model=ConfigResponse)
 async def get_config(config_manager: ConfigManager = Depends(get_config_manager)):
     """Retrieve the current system configuration."""
     return await config_manager.get_config()
 
+@router.get("/llm", response_model=Dict[str, Any])
+async def get_llm_config(config_manager: ConfigManager = Depends(get_config_manager)):
+    """Get LLM provider and model configuration."""
+    llm_config = await config_manager.get_llm_config()
+    return llm_config
+
+@router.put("/llm", response_model=Dict[str, Any])
+async def update_llm_config(
+    llm_update: LLMConfigUpdate,
+    config_manager: ConfigManager = Depends(get_config_manager)
+):
+    """Update LLM provider and model configuration."""
+    import json
+    import os
+    
+    provider = llm_update.provider
+    model = llm_update.model
+    
+    # Load base config
+    project_root = os.path.abspath(os.path.join(os.path.dirname(__file__), "..", "..", ".."))
+    config_path = os.path.join(project_root, "config", "config.json")
+    
+    base_config = {}
+    if os.path.exists(config_path):
+        try:
+            with open(config_path, "r", encoding="utf-8") as f:
+                base_config = json.load(f)
+        except Exception:
+            pass
+    
+    # Update llm config
+    if "llm" not in base_config:
+        base_config["llm"] = {}
+    
+    base_config["llm"]["provider"] = provider
+    if model:
+        # Update model in the provider-specific config
+        if provider in base_config["llm"]:
+            if not isinstance(base_config["llm"][provider], dict):
+                base_config["llm"][provider] = {}
+            base_config["llm"][provider]["model"] = model
+    
+    # Save to file
+    try:
+        with open(config_path, "w", encoding="utf-8") as f:
+            json.dump(base_config, f, ensure_ascii=False, indent=2)
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Failed to save config: {str(e)}")
+    
+    # Also update in database - use same error handling as _load_dynamic_config_from_db
+    try:
+        from app.models.system_config import SystemConfig
+        from sqlalchemy.future import select
+        
+        result = await config_manager.db.execute(
+            select(SystemConfig).where(SystemConfig.config_key == "llm")
+        )
+        existing_llm_config = result.scalar_one_or_none()
+        
+        if existing_llm_config:
+            if isinstance(existing_llm_config.config_value, dict):
+                existing_llm_config.config_value["provider"] = provider
+                if model:
+                    if provider not in existing_llm_config.config_value:
+                        existing_llm_config.config_value[provider] = {}
+                    elif not isinstance(existing_llm_config.config_value[provider], dict):
+                        existing_llm_config.config_value[provider] = {}
+                    existing_llm_config.config_value[provider]["model"] = model
+            else:
+                existing_llm_config.config_value = {"provider": provider}
+                if model:
+                    existing_llm_config.config_value[provider] = {"model": model}
+        else:
+            new_llm_config = SystemConfig(
+                config_key="llm", 
+                config_value={"provider": provider}
+            )
+            if model:
+                new_llm_config.config_value[provider] = {"model": model}
+            config_manager.db.add(new_llm_config)
+        
+        await config_manager.db.commit()
+    except Exception as e:
+        # Rollback on error
+        try:
+            await config_manager.db.rollback()
+        except Exception:
+            pass
+        
+        # Log the error but don't fail the request since file was already saved
+        import logging
+        logger = logging.getLogger(__name__)
+        logger.warning(f"Failed to update LLM config in database (file saved successfully): {e}")
+        # Continue anyway since file config was saved successfully
+    
+    return await config_manager.get_llm_config()
+
 @router.put("/", response_model=ConfigResponse)
 async def update_config(
     config_update: ConfigUpdateRequest,

backend/app/routers/financial.py

@@ -8,7 +8,7 @@ from datetime import datetime, timezone, timedelta
 from enum import Enum
 from typing import Dict, List
 
-from fastapi import APIRouter, HTTPException, Query
+from fastapi import APIRouter, HTTPException, Query, Depends
 from fastapi.responses import StreamingResponse
 
 from app.core.config import settings
@@ -24,6 +24,9 @@ from app.schemas.financial import (
 )
 from app.services.company_profile_client import CompanyProfileClient
 from app.services.analysis_client import AnalysisClient, load_analysis_config, get_analysis_config
+from app.core.dependencies import get_config_manager
+from app.services.config_manager import ConfigManager
+from app.services.client_factory import create_analysis_client
 
 # Lazy DataManager loader to avoid import-time failures when optional providers/config are missing
 _dm = None
@@ -92,6 +95,7 @@ async def get_data_sources():
 async def generate_full_analysis(
     ts_code: str,
     company_name: str = Query(None, description="Company name for better context"),
+    config_manager: ConfigManager = Depends(get_config_manager),
 ):
     """
     Generate a full analysis report by orchestrating multiple analysis modules
@@ -102,20 +106,11 @@ async def generate_full_analysis(
 
     logger.info(f"[API] Full analysis requested for {ts_code}")
 
-    # Load base and analysis configurations
-    base_cfg = _load_json(BASE_CONFIG_PATH)
-    llm_provider = base_cfg.get("llm", {}).get("provider", "gemini")
-    llm_config = base_cfg.get("llm", {}).get(llm_provider, {})
-    
-    api_key = llm_config.get("api_key")
-    base_url = llm_config.get("base_url")
-
-    if not api_key:
-        logger.error(f"[API] API key for {llm_provider} not configured")
-        raise HTTPException(
-            status_code=500,
-            detail=f"API key for {llm_provider} not configured."
-        )
+    # Load LLM configuration using ConfigManager
+    llm_config_result = await config_manager.get_llm_config()
+    default_provider = llm_config_result["provider"]
+    default_config = llm_config_result["config"]
+    global_model = llm_config_result.get("model")  # 全局模型配置
 
     analysis_config_full = load_analysis_config()
     modules_config = analysis_config_full.get("analysis_modules", {})
@@ -211,10 +206,15 @@ async def generate_full_analysis(
         module_config = modules_config[module_type]
         logger.info(f"[Orchestrator] Starting analysis for module: {module_type}")
         
-        client = AnalysisClient(
-            api_key=api_key, 
-            base_url=base_url, 
-            model=module_config.get("model", "gemini-1.5-flash")
+        # 统一使用全局配置，不再从模块配置读取 provider 和 model
+        # 使用全局 provider 和 model
+        model = global_model or default_config.get("model", "gemini-1.5-flash")
+        
+        # Create client using factory with global config
+        client = create_analysis_client(
+            provider=default_provider,
+            config=default_config,
+            model=model
         )
         
         # Gather context from completed dependencies
@@ -468,6 +468,7 @@ async def get_financials(
 async def get_company_profile(
     ts_code: str,
     company_name: str = Query(None, description="Company name for better context"),
+    config_manager: ConfigManager = Depends(get_config_manager),
 ):
     """
     Get company profile for a company using Gemini AI (non-streaming, single response)
@@ -477,19 +478,26 @@ async def get_company_profile(
     
     logger.info(f"[API] Company profile requested for {ts_code}")
     
-    # Load config
-    base_cfg = _load_json(BASE_CONFIG_PATH)
-    llm_provider = base_cfg.get("llm", {}).get("provider", "gemini")
-    llm_config = base_cfg.get("llm", {}).get(llm_provider, {})
+    # Load LLM configuration using ConfigManager
+    llm_config_result = await config_manager.get_llm_config()
+    provider = llm_config_result["provider"]
+    provider_config = llm_config_result["config"]
     
-    api_key = llm_config.get("api_key")
-    base_url = llm_config.get("base_url") # Will be None if not set, handled by client
+    # CompanyProfileClient only supports OpenAI-compatible APIs
+    if provider == "alpha_engine":
+        raise HTTPException(
+            status_code=400,
+            detail="Company profile generation does not support AlphaEngine provider. Please use OpenAI-compatible API."
+        )
+    
+    api_key = provider_config.get("api_key")
+    base_url = provider_config.get("base_url")
     
     if not api_key:
-        logger.error(f"[API] API key for {llm_provider} not configured")
+        logger.error(f"[API] API key for {provider} not configured")
         raise HTTPException(
             status_code=500, 
-            detail=f"API key for {llm_provider} not configured."
+            detail=f"API key for {provider} not configured."
         )
     
     client = CompanyProfileClient(
@@ -573,6 +581,7 @@ async def generate_analysis(
     ts_code: str,
     analysis_type: str,
     company_name: str = Query(None, description="Company name for better context"),
+    config_manager: ConfigManager = Depends(get_config_manager),
 ):
     """
     Generate analysis for a company using Gemini AI
@@ -591,37 +600,11 @@ async def generate_analysis(
     
     logger.info(f"[API] Analysis requested for {ts_code}, type: {analysis_type}")
     
-    # Load config
-    base_cfg = _load_json(BASE_CONFIG_PATH)
-    llm_provider = base_cfg.get("llm", {}).get("provider", "gemini")
-    llm_config = base_cfg.get("llm", {}).get(llm_provider, {})
-    
-    api_key = llm_config.get("api_key")
-    base_url = llm_config.get("base_url")
-
-    if not api_key:
-        logger.error(f"[API] API key for {llm_provider} not configured")
-        raise HTTPException(
-            status_code=500, 
-            detail=f"API key for {llm_provider} not configured."
-        )
-    
-    # Get analysis configuration
-    analysis_cfg = get_analysis_config(analysis_type)
-    if not analysis_cfg:
-        raise HTTPException(
-            status_code=404,
-            detail=f"Analysis type '{analysis_type}' not found in configuration"
-        )
-    
-    model = analysis_cfg.get("model", "gemini-2.5-flash")
-    prompt_template = analysis_cfg.get("prompt_template", "")
-    
-    if not prompt_template:
-        raise HTTPException(
-            status_code=500,
-            detail=f"Prompt template not found for analysis type '{analysis_type}'"
-        )
+    # Load LLM configuration using ConfigManager
+    llm_config_result = await config_manager.get_llm_config()
+    default_provider = llm_config_result["provider"]
+    default_config = llm_config_result["config"]
+    global_model = llm_config_result.get("model")  # 全局模型配置
     
     # Get company name from ts_code if not provided
     financial_data = None
@@ -656,8 +639,30 @@ async def generate_analysis(
     
     logger.info(f"[API] Generating {analysis_type} for {company_name}")
     
-    # Initialize analysis client with configured model
-    client = AnalysisClient(api_key=api_key, base_url=base_url, model=model)
+    # Get analysis configuration for prompt template
+    analysis_cfg = get_analysis_config(analysis_type)
+    if not analysis_cfg:
+        raise HTTPException(
+            status_code=404,
+            detail=f"Analysis type '{analysis_type}' not found in configuration"
+        )
+    
+    prompt_template = analysis_cfg.get("prompt_template", "")
+    if not prompt_template:
+        raise HTTPException(
+            status_code=500,
+            detail=f"Prompt template not found for analysis type '{analysis_type}'"
+        )
+    
+    # 统一使用全局配置，不再从模块配置读取 provider 和 model
+    model = global_model or default_config.get("model", "gemini-1.5-flash")
+    
+    # 统一使用全局配置创建客户端
+    client = create_analysis_client(
+        provider=default_provider,
+        config=default_config,
+        model=model
+    )
     
     # Prepare dependency context for single-module generation
     # If the requested module declares dependencies, generate them first and inject their outputs
@@ -701,12 +706,18 @@ async def generate_analysis(
                 # Fallback: if cycle detected, just use any order
                 order = list(all_required)
 
-            # Generate dependencies in order
+            # Generate dependencies in order - 统一使用全局配置
             completed = {}
             for mod in order:
                 cfg = modules_config.get(mod, {})
                 dep_ctx = {d: completed.get(d, "") for d in (cfg.get("dependencies", []) or [])}
-                dep_client = AnalysisClient(api_key=api_key, base_url=base_url, model=cfg.get("model", model))
+                
+                # 统一使用全局配置，不再从模块配置读取
+                dep_client = create_analysis_client(
+                    provider=default_provider,
+                    config=default_config,
+                    model=model
+                )
                 dep_result = await dep_client.generate_analysis(
                     analysis_type=mod,
                     company_name=company_name,
@@ -888,6 +899,7 @@ async def stream_analysis(
     ts_code: str,
     analysis_type: str,
     company_name: str = Query(None, description="Company name for better context"),
+    config_manager: ConfigManager = Depends(get_config_manager),
 ):
     """
     Stream analysis content chunks for a given module using OpenAI-compatible streaming.
@@ -899,24 +911,19 @@ async def stream_analysis(
 
     logger.info(f"[API] Streaming analysis requested for {ts_code}, type: {analysis_type}")
 
-    # Load config
-    base_cfg = _load_json(BASE_CONFIG_PATH)
-    llm_provider = base_cfg.get("llm", {}).get("provider", "gemini")
-    llm_config = base_cfg.get("llm", {}).get(llm_provider, {})
-
-    api_key = llm_config.get("api_key")
-    base_url = llm_config.get("base_url")
-
-    if not api_key:
-        logger.error(f"[API] API key for {llm_provider} not configured")
-        raise HTTPException(status_code=500, detail=f"API key for {llm_provider} not configured.")
+    # Load LLM configuration using ConfigManager
+    llm_config_result = await config_manager.get_llm_config()
+    default_provider = llm_config_result["provider"]
+    default_config = llm_config_result["config"]
+    global_model = llm_config_result.get("model")  # 全局模型配置
 
     # Get analysis configuration
     analysis_cfg = get_analysis_config(analysis_type)
     if not analysis_cfg:
         raise HTTPException(status_code=404, detail=f"Analysis type '{analysis_type}' not found in configuration")
 
-    model = analysis_cfg.get("model", "gemini-2.5-flash")
+    # 统一使用全局配置，不再从模块配置读取 provider 和 model
+    model = global_model or default_config.get("model", "gemini-1.5-flash")
     prompt_template = analysis_cfg.get("prompt_template", "")
     if not prompt_template:
         raise HTTPException(status_code=500, detail=f"Prompt template not found for analysis type '{analysis_type}'")
@@ -972,7 +979,13 @@ async def stream_analysis(
             for mod in order:
                 cfg = modules_config.get(mod, {})
                 dep_ctx = {d: completed.get(d, "") for d in (cfg.get("dependencies", []) or [])}
-                dep_client = AnalysisClient(api_key=api_key, base_url=base_url, model=cfg.get("model", model))
+                
+                # 统一使用全局配置，不再从模块配置读取
+                dep_client = create_analysis_client(
+                    provider=default_provider,
+                    config=default_config,
+                    model=model
+                )
                 dep_result = await dep_client.generate_analysis(
                     analysis_type=mod,
                     company_name=company_name,
@@ -986,7 +999,12 @@ async def stream_analysis(
     except Exception:
         context = {}
 
-    client = AnalysisClient(api_key=api_key, base_url=base_url, model=model)
+    # 统一使用全局配置创建客户端
+    client = create_analysis_client(
+        provider=default_provider,
+        config=default_config,
+        model=model
+    )
 
     async def streamer():
         # Optional header line to help client-side UI

backend/app/routers/orgs.py

@@ -2,7 +2,7 @@ import logging
 import os
 import json
 from typing import Dict
-from fastapi import APIRouter, BackgroundTasks, HTTPException
+from fastapi import APIRouter, BackgroundTasks, HTTPException, Depends
 
 # Lazy loader for DataManager
 _dm = None
@@ -23,6 +23,9 @@ def get_dm():
         return _dm
 
 from app.services.analysis_client import AnalysisClient, load_analysis_config
+from app.core.dependencies import get_config_manager
+from app.services.config_manager import ConfigManager
+from app.services.client_factory import create_analysis_client
 
 router = APIRouter()
 logger = logging.getLogger(__name__)
@@ -40,7 +43,7 @@ def _load_json(path: str) -> Dict:
     except Exception:
         return {}
 
-async def run_full_analysis(org_id: str):
+async def run_full_analysis(org_id: str, config_manager: ConfigManager = None):
     """
     Asynchronous task to run a full analysis for a given stock.
     This function is market-agnostic and relies on DataManager.
@@ -48,16 +51,23 @@ async def run_full_analysis(org_id: str):
     logger.info(f"Starting full analysis task for {org_id}")
 
     # 1. Load configurations
-    base_cfg = _load_json(BASE_CONFIG_PATH)
-    llm_provider = base_cfg.get("llm", {}).get("provider", "gemini")
-    llm_config = base_cfg.get("llm", {}).get(llm_provider, {})
+    if config_manager is None:
+        # If called from background task, we need to create a new session
+        from app.core.database import AsyncSessionLocal
+        async with AsyncSessionLocal() as session:
+            config_manager = ConfigManager(db_session=session)
+            await _run_analysis_with_config(org_id, config_manager)
+    else:
+        await _run_analysis_with_config(org_id, config_manager)
 
-    api_key = llm_config.get("api_key")
-    base_url = llm_config.get("base_url")
 
-    if not api_key:
-        logger.error(f"API key for {llm_provider} not configured. Aborting analysis for {org_id}.")
-        return
+async def _run_analysis_with_config(org_id: str, config_manager: ConfigManager):
+    """Internal function to run analysis with a ConfigManager instance"""
+    # Load LLM configuration using ConfigManager
+    llm_config_result = await config_manager.get_llm_config()
+    default_provider = llm_config_result["provider"]
+    default_config = llm_config_result["config"]
+    global_model = llm_config_result.get("model")  # 全局模型配置
 
     analysis_config_full = load_analysis_config()
     modules_config = analysis_config_full.get("analysis_modules", {})
@@ -96,10 +106,15 @@ async def run_full_analysis(org_id: str):
     analysis_results = {}
     for module_type, module_config in modules_config.items():
         logger.info(f"Running analysis module: {module_type} for {org_id}")
-        client = AnalysisClient(
-            api_key=api_key, 
-            base_url=base_url, 
-            model=module_config.get("model", "gemini-1.5-flash")
+        
+        # 统一使用全局配置，不再从模块配置读取 provider 和 model
+        model = global_model or default_config.get("model", "gemini-1.5-flash")
+        
+        # Create client using factory with global config
+        client = create_analysis_client(
+            provider=default_provider,
+            config=default_config,
+            model=model
         )
         
         # Simplified context: use results from all previously completed modules
@@ -128,7 +143,7 @@ async def run_full_analysis(org_id: str):
 
 
 @router.post("/{market}/{org_id}/reports/generate")
-async def trigger_report_generation(market: str, org_id: str, background_tasks: BackgroundTasks):
+async def trigger_report_generation(market: str, org_id: str, background_tasks: BackgroundTasks, config_manager: ConfigManager = Depends(get_config_manager)):
     """
     Triggers a background task to generate a full financial report.
     This endpoint is now market-agnostic.
@@ -137,7 +152,8 @@ async def trigger_report_generation(market: str, org_id: str, background_tasks:
     
     # TODO: Create a report record in the database with "generating" status here.
     
-    background_tasks.add_task(run_full_analysis, org_id)
+    # Pass config_manager to the background task
+    background_tasks.add_task(run_full_analysis, org_id, config_manager)
     
     logger.info(f"Queued analysis task for {org_id}.")
     return {"queued": True, "market": market, "org_id": org_id}

backend/app/schemas/config.py

@@ -1,7 +1,7 @@
 """
 Configuration-related Pydantic schemas
 """
-from typing import Dict, Optional, Any
+from typing import Dict, Optional, Any, List
 from pydantic import BaseModel, Field
 
 class DatabaseConfig(BaseModel):
@@ -11,17 +11,30 @@ class NewApiConfig(BaseModel):
     api_key: str = Field(..., description="New API Key")
     base_url: Optional[str] = None
 
+class AlphaEngineConfig(BaseModel):
+    api_url: str = Field(..., description="AlphaEngine API URL")
+    api_key: str = Field(..., description="AlphaEngine API Key")
+    token: str = Field(..., description="AlphaEngine Token")
+    user_id: int = Field(999041, description="User ID")
+    model: str = Field("deepseek-r1", description="Model name")
+    using_indicator: bool = Field(True, description="Whether to use indicators")
+    start_time: str = Field("2024-01-01", description="Start time for data query")
+    doc_show_type: List[str] = Field(["A001", "A002", "A003", "A004"], description="Document types")
+    simple_tracking: bool = Field(True, description="Whether to enable simple tracking")
+
 class DataSourceConfig(BaseModel):
     api_key: str = Field(..., description="数据源API Key")
 
 class ConfigResponse(BaseModel):
     database: DatabaseConfig
     new_api: NewApiConfig
+    alpha_engine: Optional[AlphaEngineConfig] = None
     data_sources: Dict[str, DataSourceConfig]
 
 class ConfigUpdateRequest(BaseModel):
     database: Optional[DatabaseConfig] = None
     new_api: Optional[NewApiConfig] = None
+    alpha_engine: Optional[AlphaEngineConfig] = None
     data_sources: Optional[Dict[str, DataSourceConfig]] = None
 
 class ConfigTestRequest(BaseModel):

backend/app/services/alpha_engine_client.py

@@ -0,0 +1,260 @@
+"""
+AlphaEngine Client for investment Q&A API
+"""
+import json
+import re
+import time
+from typing import Dict, Optional, AsyncGenerator
+import httpx
+from requests.exceptions import ChunkedEncodingError
+
+
+class AlphaEngineClient:
+    """Client for AlphaEngine investment Q&A API"""
+    
+    def __init__(
+        self,
+        api_url: str,
+        api_key: str,
+        token: str,
+        user_id: int = 999041,
+        model: str = "deepseek-r1",
+        using_indicator: bool = True,
+        start_time: str = "2024-01-01",
+        doc_show_type: list = None,
+        simple_tracking: bool = True
+    ):
+        """
+        Initialize AlphaEngine client
+        
+        Args:
+            api_url: API endpoint URL
+            api_key: X-API-KEY for authentication
+            token: Token for authentication
+            user_id: User ID
+            model: Model name (default: deepseek-r1)
+            using_indicator: Whether to use indicators
+            start_time: Start time for data query
+            doc_show_type: Document types to show (default: ["A001", "A002", "A003", "A004"])
+            simple_tracking: Whether to enable simple tracking
+        """
+        self.api_url = api_url.rstrip('/')
+        self.api_key = api_key
+        self.token = token
+        self.user_id = user_id
+        self.model = model
+        self.using_indicator = using_indicator
+        self.start_time = start_time
+        self.doc_show_type = doc_show_type or ["A001", "A002", "A003", "A004"]
+        self.simple_tracking = simple_tracking
+    
+    async def generate_analysis(
+        self,
+        analysis_type: str,
+        company_name: str,
+        ts_code: str,
+        prompt_template: str,
+        financial_data: Optional[Dict] = None,
+        context: Optional[Dict] = None
+    ) -> Dict:
+        """
+        Generate analysis using AlphaEngine API (non-streaming)
+        
+        Args:
+            analysis_type: Type of analysis
+            company_name: Company name
+            ts_code: Stock code
+            prompt_template: Prompt template with placeholders
+            financial_data: Optional financial data for context
+            context: Optional dictionary with results from previous analyses
+            
+        Returns:
+            Dict with analysis content and metadata
+        """
+        start_time = time.perf_counter_ns()
+        
+        # Build prompt from template
+        prompt = self._build_prompt(
+            prompt_template,
+            company_name,
+            ts_code,
+            financial_data,
+            context
+        )
+        
+        # Call AlphaEngine API
+        try:
+            async with httpx.AsyncClient(timeout=300.0) as client:
+                headers = {
+                    'token': self.token,
+                    'X-API-KEY': self.api_key,
+                    'Content-Type': 'application/json'
+                }
+                
+                payload = {
+                    "msg": prompt,
+                    "history": [],
+                    "user_id": self.user_id,
+                    "model": self.model,
+                    "using_indicator": self.using_indicator,
+                    "start_time": self.start_time,
+                    "doc_show_type": self.doc_show_type,
+                    "simple_tracking": self.simple_tracking
+                }
+                
+                response = await client.post(
+                    f"{self.api_url}/api/v3/finchat",
+                    json=payload,
+                    headers=headers
+                )
+                
+                if response.status_code != 200:
+                    raise Exception(f"AlphaEngine API error: HTTP {response.status_code} - {response.text}")
+                
+                result_text = response.text
+                
+                # Parse response to extract final answer
+                final_answer_match = re.findall(r'\{"id":"_final","content":"(.*?)"}', result_text)
+                final_answer = final_answer_match[0] if final_answer_match else result_text
+                
+                # Extract COT if available
+                cot_match = re.findall(r'\{"id":"_cot","content":"(.*?)"}', result_text)
+                cot = "".join(cot_match) if cot_match else ""
+                
+                # Extract tracking documents if available
+                tracking_match = re.findall(r'\{"id":"tracking_documents","content":\s*(\[[^]]*])}', result_text)
+                tracking_docs = json.loads(tracking_match[0]) if tracking_match else []
+                
+                elapsed_ms = int((time.perf_counter_ns() - start_time) / 1_000_000)
+                
+                return {
+                    "content": final_answer,
+                    "model": self.model,
+                    "tokens": {
+                        "prompt_tokens": 0,  # AlphaEngine doesn't provide token usage
+                        "completion_tokens": 0,
+                        "total_tokens": 0,
+                    },
+                    "elapsed_ms": elapsed_ms,
+                    "success": True,
+                    "analysis_type": analysis_type,
+                    "cot": cot,
+                    "tracking_documents": tracking_docs,
+                }
+        except Exception as e:
+            elapsed_ms = int((time.perf_counter_ns() - start_time) / 1_000_000)
+            return {
+                "content": "",
+                "model": self.model,
+                "tokens": {"prompt_tokens": 0, "completion_tokens": 0, "total_tokens": 0},
+                "elapsed_ms": elapsed_ms,
+                "success": False,
+                "error": str(e),
+                "analysis_type": analysis_type,
+            }
+    
+    async def generate_analysis_stream(
+        self,
+        analysis_type: str,
+        company_name: str,
+        ts_code: str,
+        prompt_template: str,
+        financial_data: Optional[Dict] = None,
+        context: Optional[Dict] = None
+    ) -> AsyncGenerator[str, None]:
+        """
+        Yield analysis content chunks using AlphaEngine streaming API
+        
+        Yields plain text chunks as they arrive.
+        """
+        # Build prompt
+        prompt = self._build_prompt(
+            prompt_template,
+            company_name,
+            ts_code,
+            financial_data,
+            context,
+        )
+        
+        try:
+            async with httpx.AsyncClient(timeout=300.0) as client:
+                headers = {
+                    'token': self.token,
+                    'X-API-KEY': self.api_key,
+                    'Content-Type': 'application/json'
+                }
+                
+                payload = {
+                    "msg": prompt,
+                    "history": [],
+                    "user_id": self.user_id,
+                    "model": self.model,
+                    "using_indicator": self.using_indicator,
+                    "start_time": self.start_time,
+                    "doc_show_type": self.doc_show_type,
+                    "simple_tracking": self.simple_tracking
+                }
+                
+                async with client.stream(
+                    "POST",
+                    f"{self.api_url}/api/v3/finchat",
+                    json=payload,
+                    headers=headers
+                ) as response:
+                    if response.status_code != 200:
+                        yield f"\n\n[错误] HTTP {response.status_code}: {response.text}\n"
+                        return
+                    
+                    async for chunk in response.aiter_bytes(chunk_size=128):
+                        try:
+                            chunk_text = chunk.decode('utf-8', 'ignore')
+                            yield chunk_text
+                        except UnicodeDecodeError:
+                            chunk_text = chunk.decode('utf-8', 'replace')
+                            yield chunk_text
+        except Exception as e:
+            yield f"\n\n[错误] {type(e).__name__}: {str(e)}\n"
+    
+    def _build_prompt(
+        self,
+        prompt_template: str,
+        company_name: str,
+        ts_code: str,
+        financial_data: Optional[Dict] = None,
+        context: Optional[Dict] = None
+    ) -> str:
+        """Build prompt from template by replacing placeholders"""
+        import string
+        
+        # Start with base placeholders
+        placeholders = {
+            "company_name": company_name,
+            "ts_code": ts_code,
+        }
+        
+        # Add financial data if provided
+        financial_data_str = ""
+        if financial_data:
+            try:
+                financial_data_str = json.dumps(financial_data, ensure_ascii=False, indent=2)
+            except Exception:
+                financial_data_str = str(financial_data)
+        placeholders["financial_data"] = financial_data_str
+        
+        # Add context from previous analysis steps
+        if context:
+            placeholders.update(context)
+        
+        # Replace placeholders in template
+        class SafeFormatter(string.Formatter):
+            def get_value(self, key, args, kwargs):
+                if isinstance(key, str):
+                    return kwargs.get(key, f"{{{key}}}")
+                else:
+                    return super().get_value(key, args, kwargs)
+        
+        formatter = SafeFormatter()
+        prompt = formatter.format(prompt_template, **placeholders)
+        
+        return prompt
+

backend/app/services/client_factory.py

@@ -0,0 +1,60 @@
+"""
+Unified Analysis Client Factory
+Creates appropriate client based on provider type
+"""
+from typing import Dict, Optional
+from app.services.analysis_client import AnalysisClient
+from app.services.alpha_engine_client import AlphaEngineClient
+
+
+def create_analysis_client(
+    provider: str,
+    config: Dict,
+    model: str = None
+):
+    """
+    Create an analysis client based on provider type
+    
+    Args:
+        provider: Provider type ("openai", "gemini", "new_api", "alpha_engine")
+        config: Configuration dictionary containing provider-specific settings
+        model: Model name (optional, may be overridden by config)
+        
+    Returns:
+        Client instance (AnalysisClient or AlphaEngineClient)
+    """
+    if provider == "alpha_engine":
+        # AlphaEngine specific configuration
+        api_url = config.get("api_url", "")
+        api_key = config.get("api_key", "")
+        token = config.get("token", "")
+        user_id = config.get("user_id", 999041)
+        model_name = model or config.get("model", "deepseek-r1")
+        using_indicator = config.get("using_indicator", True)
+        start_time = config.get("start_time", "2024-01-01")
+        doc_show_type = config.get("doc_show_type", ["A001", "A002", "A003", "A004"])
+        simple_tracking = config.get("simple_tracking", True)
+        
+        return AlphaEngineClient(
+            api_url=api_url,
+            api_key=api_key,
+            token=token,
+            user_id=user_id,
+            model=model_name,
+            using_indicator=using_indicator,
+            start_time=start_time,
+            doc_show_type=doc_show_type,
+            simple_tracking=simple_tracking
+        )
+    else:
+        # OpenAI-compatible API (openai, gemini, new_api)
+        api_key = config.get("api_key", "")
+        base_url = config.get("base_url", "")
+        model_name = model or config.get("model", "gemini-1.5-flash")
+        
+        return AnalysisClient(
+            api_key=api_key,
+            base_url=base_url,
+            model=model_name
+        )
+

backend/app/services/company_profile_client.py

@@ -68,14 +68,14 @@ class CompanyProfileClient:
                 "error": str(e),
             }
     
-    def generate_profile_stream(
+    async def generate_profile_stream(
         self, 
         company_name: str, 
         ts_code: str,
         financial_data: Optional[Dict] = None
     ):
         """
-        Generate company profile using Gemini API with streaming
+        Generate company profile using OpenAI-compatible streaming API
         
         Args:
             company_name: Company name
@@ -85,40 +85,31 @@ class CompanyProfileClient:
         Yields:
             Chunks of generated content
         """
-        import logging
-        logger = logging.getLogger(__name__)
-        
-        logger.info(f"[CompanyProfile] Starting stream generation for {company_name} ({ts_code})")
-        
         # Build prompt
         prompt = self._build_prompt(company_name, ts_code, financial_data)
-        logger.info(f"[CompanyProfile] Prompt built, length: {len(prompt)} chars")
         
-        # Call Gemini API with streaming
+        # Call OpenAI-compatible API with streaming
         try:
-            logger.info("[CompanyProfile] Calling Gemini API with stream=True")
-            # Generate streaming response (sync call, but yields chunks)
-            response_stream = self.model.generate_content(prompt, stream=True)
-            logger.info("[CompanyProfile] Gemini API stream object created")
-            
-            chunk_count = 0
-            # Stream chunks
-            logger.info("[CompanyProfile] Starting to iterate response stream")
-            for chunk in response_stream:
-                logger.info(f"[CompanyProfile] Received chunk from Gemini, has text: {hasattr(chunk, 'text')}")
-                if hasattr(chunk, 'text') and chunk.text:
-                    chunk_count += 1
-                    text_len = len(chunk.text)
-                    logger.info(f"[CompanyProfile] Chunk {chunk_count}: {text_len} chars")
-                    yield chunk.text
-                else:
-                    logger.warning(f"[CompanyProfile] Chunk has no text attribute or empty, chunk: {chunk}")
-            
-            logger.info(f"[CompanyProfile] Stream iteration completed. Total chunks: {chunk_count}")
+            stream = await self.client.chat.completions.create(
+                model=self.model_name,
+                messages=[{"role": "user", "content": prompt}],
+                stream=True,
+            )
             
+            # The SDK yields events with incremental deltas
+            async for event in stream:
+                try:
+                    choice = event.choices[0] if getattr(event, "choices", None) else None
+                    delta = getattr(choice, "delta", None) if choice is not None else None
+                    content = getattr(delta, "content", None) if delta is not None else None
+                    if content:
+                        yield content
+                except Exception:
+                    # Best-effort: ignore malformed chunks
+                    continue
         except Exception as e:
-            logger.error(f"[CompanyProfile] Error during streaming: {type(e).__name__}: {str(e)}", exc_info=True)
-            yield f"\n\n---\n\n**错误**: {type(e).__name__}: {str(e)}"
+            # Emit error message to the stream so the client can surface it
+            yield f"\n\n[错误] {type(e).__name__}: {str(e)}\n"
     
     def _build_prompt(self, company_name: str, ts_code: str, financial_data: Optional[Dict] = None) -> str:
         """Build prompt for company profile generation"""

backend/app/services/config_manager.py

@@ -12,7 +12,7 @@ from sqlalchemy.ext.asyncio import AsyncSession
 from sqlalchemy.future import select
 
 from app.models.system_config import SystemConfig
-from app.schemas.config import ConfigResponse, ConfigUpdateRequest, DatabaseConfig, NewApiConfig, DataSourceConfig, ConfigTestResponse
+from app.schemas.config import ConfigResponse, ConfigUpdateRequest, DatabaseConfig, NewApiConfig, AlphaEngineConfig, DataSourceConfig, ConfigTestResponse
 
 class ConfigManager:
     """Manages system configuration by merging a static JSON file with dynamic settings from the database."""
@@ -72,25 +72,114 @@ class ConfigManager:
         # 兼容两种位置：优先使用 new_api，其次回退到 llm.new_api
         new_api_src = merged_config.get("new_api") or merged_config.get("llm", {}).get("new_api", {})
         
+        # 获取 alpha_engine 配置
+        alpha_engine_src = merged_config.get("alpha_engine") or merged_config.get("llm", {}).get("alpha_engine")
+        alpha_engine_config = None
+        if alpha_engine_src:
+            alpha_engine_config = AlphaEngineConfig(**alpha_engine_src)
+
         return ConfigResponse(
             database=DatabaseConfig(**merged_config.get("database", {})),
             new_api=NewApiConfig(**(new_api_src or {})),
+            alpha_engine=alpha_engine_config,
             data_sources={
                 k: DataSourceConfig(**v) 
                 for k, v in merged_config.get("data_sources", {}).items()
             }
         )
     
+    async def get_llm_config(self, provider: str = None) -> Dict[str, Any]:
+        """
+        Get LLM configuration for a specific provider
+        
+        Args:
+            provider: Provider name (e.g., "new_api", "gemini", "alpha_engine")
+                     If None, uses the configured provider from config
+            
+        Returns:
+            Dictionary with provider configuration and provider name
+        """
+        base_config = self._load_base_config_from_file()
+        db_config = await self._load_dynamic_config_from_db()
+        
+        merged_config = self._merge_configs(base_config, db_config)
+        
+        llm_config = merged_config.get("llm", {})
+        
+        # Determine provider
+        if not provider:
+            provider = llm_config.get("provider", "new_api")
+        
+        # Get provider-specific config
+        provider_config = llm_config.get(provider, {})
+        
+        # Get global model from provider config if available
+        global_model = provider_config.get("model")
+        
+        return {
+            "provider": provider,
+            "config": provider_config,
+            "model": global_model  # 返回全局模型配置
+        }
+
+    def _filter_empty_values(self, config_dict: Dict[str, Any]) -> Dict[str, Any]:
+        """Remove empty strings and None values from config dict, but keep 0 and False."""
+        filtered = {}
+        for key, value in config_dict.items():
+            if isinstance(value, dict):
+                filtered_value = self._filter_empty_values(value)
+                if filtered_value:  # Only add if dict is not empty
+                    filtered[key] = filtered_value
+            elif value is not None and value != "":
+                filtered[key] = value
+        return filtered
+
     async def update_config(self, config_update: ConfigUpdateRequest) -> ConfigResponse:
         """Updates configuration in the database and returns the new merged config."""
         try:
             update_dict = config_update.dict(exclude_unset=True)
             
+            # 过滤空值
+            update_dict = self._filter_empty_values(update_dict)
+            
             # 验证配置数据
             self._validate_config_data(update_dict)
             
+            # 处理 LLM 相关配置：需要保存到 llm 配置下
+            llm_updates = {}
+            if "new_api" in update_dict:
+                llm_updates["new_api"] = update_dict.pop("new_api")
+            if "alpha_engine" in update_dict:
+                llm_updates["alpha_engine"] = update_dict.pop("alpha_engine")
+            
+            # 保存 LLM 配置
+            if llm_updates:
+                result = await self.db.execute(
+                    select(SystemConfig).where(SystemConfig.config_key == "llm")
+                )
+                existing_llm_config = result.scalar_one_or_none()
+                
+                if existing_llm_config:
+                    if isinstance(existing_llm_config.config_value, dict):
+                        merged_llm = self._merge_configs(existing_llm_config.config_value, llm_updates)
+                        existing_llm_config.config_value = merged_llm
+                    else:
+                        existing_llm_config.config_value = llm_updates
+                else:
+                    # 从文件加载基础配置，然后合并
+                    base_config = self._load_base_config_from_file()
+                    base_llm = base_config.get("llm", {})
+                    merged_llm = self._merge_configs(base_llm, llm_updates)
+                    new_llm_config = SystemConfig(config_key="llm", config_value=merged_llm)
+                    self.db.add(new_llm_config)
+            
+            # 保存其他配置（database, data_sources 等）
             for key, value in update_dict.items():
-                existing_config = await self.db.get(SystemConfig, key)
+                result = await self.db.execute(
+                    select(SystemConfig).where(SystemConfig.config_key == key)
+                )
+                existing_config = result.scalar_one_or_none()
+                
                 if existing_config:
                     # Merge with existing DB value before updating
                     if isinstance(existing_config.config_value, dict) and isinstance(value, dict):
@@ -112,23 +201,32 @@ class ConfigManager:
         """Validate configuration data before saving."""
         if "database" in config_data:
             db_config = config_data["database"]
-            if "url" in db_config:
+            if "url" in db_config and db_config["url"]:
                 url = db_config["url"]
                 if not url.startswith(("postgresql://", "postgresql+asyncpg://")):
                     raise ValueError("数据库URL必须以 postgresql:// 或 postgresql+asyncpg:// 开头")
         
         if "new_api" in config_data:
             new_api_config = config_data["new_api"]
-            if "api_key" in new_api_config and len(new_api_config["api_key"]) < 10:
+            if "api_key" in new_api_config and new_api_config["api_key"] and len(new_api_config["api_key"]) < 10:
                 raise ValueError("New API Key长度不能少于10个字符")
             if "base_url" in new_api_config and new_api_config["base_url"]:
                 base_url = new_api_config["base_url"]
                 if not base_url.startswith(("http://", "https://")):
                     raise ValueError("New API Base URL必须以 http:// 或 https:// 开头")
         
+        if "alpha_engine" in config_data:
+            alpha_engine_config = config_data["alpha_engine"]
+            if "api_key" in alpha_engine_config and alpha_engine_config["api_key"] and len(alpha_engine_config["api_key"]) < 5:
+                raise ValueError("AlphaEngine API Key长度不能少于5个字符")
+            if "api_url" in alpha_engine_config and alpha_engine_config["api_url"]:
+                api_url = alpha_engine_config["api_url"]
+                if not api_url.startswith(("http://", "https://")):
+                    raise ValueError("AlphaEngine API URL必须以 http:// 或 https:// 开头")
+        
         if "data_sources" in config_data:
             for source_name, source_config in config_data["data_sources"].items():
-                if "api_key" in source_config and len(source_config["api_key"]) < 10:
+                if "api_key" in source_config and source_config["api_key"] and len(source_config["api_key"]) < 10:
                     raise ValueError(f"{source_name} API Key长度不能少于10个字符")
 
     async def test_config(self, config_type: str, config_data: Dict[str, Any]) -> ConfigTestResponse:
@@ -142,6 +240,8 @@ class ConfigManager:
                 return await self._test_tushare(config_data)
             elif config_type == "finnhub":
                 return await self._test_finnhub(config_data)
+            elif config_type == "alpha_engine":
+                return await self._test_alpha_engine(config_data)
             else:
                 return ConfigTestResponse(
                     success=False,
@@ -302,3 +402,57 @@ class ConfigManager:
                 success=False,
                 message=f"Finnhub API连接失败: {str(e)}"
             )
+    
+    async def _test_alpha_engine(self, config_data: Dict[str, Any]) -> ConfigTestResponse:
+        """Test AlphaEngine API connection."""
+        api_url = config_data.get("api_url")
+        api_key = config_data.get("api_key")
+        token = config_data.get("token")
+        
+        if not api_url or not api_key or not token:
+            return ConfigTestResponse(
+                success=False,
+                message="AlphaEngine API URL、API Key和Token均不能为空"
+            )
+        
+        try:
+            async with httpx.AsyncClient(timeout=10.0) as client:
+                headers = {
+                    'token': token,
+                    'X-API-KEY': api_key,
+                    'Content-Type': 'application/json'
+                }
+                
+                # 发送一个简单的测试请求
+                payload = {
+                    "msg": "测试连接",
+                    "history": [],
+                    "user_id": config_data.get("user_id", 999041),
+                    "model": config_data.get("model", "deepseek-r1"),
+                    "using_indicator": config_data.get("using_indicator", True),
+                    "start_time": config_data.get("start_time", "2024-01-01"),
+                    "doc_show_type": config_data.get("doc_show_type", ["A001", "A002", "A003", "A004"]),
+                    "simple_tracking": config_data.get("simple_tracking", True)
+                }
+                
+                response = await client.post(
+                    f"{api_url.rstrip('/')}/api/v3/finchat",
+                    json=payload,
+                    headers=headers
+                )
+                
+                if response.status_code == 200:
+                    return ConfigTestResponse(
+                        success=True,
+                        message="AlphaEngine API连接成功"
+                    )
+                else:
+                    return ConfigTestResponse(
+                        success=False,
+                        message=f"AlphaEngine API测试失败: HTTP {response.status_code} - {response.text[:200]}"
+                    )
+        except Exception as e:
+            return ConfigTestResponse(
+                success=False,
+                message=f"AlphaEngine API连接失败: {str(e)}"
+            )

config/config.json

@@ -1,6 +1,6 @@
 {
   "llm": {
-    "provider": "new_api",
+    "provider": "alpha_engine",
     "gemini": {
       "base_url": "",
       "api_key": "YOUR_GEMINI_API_KEY"
@@ -8,6 +8,22 @@
     "new_api": {
       "base_url": "http://192.168.3.214:3000/v1",
       "api_key": "sk-DdTTQ5fdU1aFW6gnYxSNYDgFsVQg938zUcmY4vaB7oPtcNs7"
+    },
+    "alpha_engine": {
+      "api_url": "http://api-ai-prod.valuesimplex.tech",
+      "api_key": "api@shangjian!",
+      "token": "9b5c0b6a5e1e4e8fioouiouqiuioasaz",
+      "user_id": 999041,
+      "model": "deepseek-r1",
+      "using_indicator": true,
+      "start_time": "2024-01-01",
+      "doc_show_type": [
+        "A001",
+        "A002",
+        "A003",
+        "A004"
+      ],
+      "simple_tracking": true
     }
   },
   "data_sources": {

docs/AlphaEngine/2 test_investment_qa_v3(1).py

@@ -0,0 +1,99 @@
+# coding:utf-8
+import json
+import re
+
+import requests
+from requests.exceptions import ChunkedEncodingError
+
+# 请求地址
+qa_url = "http://api-ai-prod.valuesimplex.tech/api/v3/finchat"
+# 熵简提供的x-api-key
+api_key = "api@shangjian!"
+token = "9b5c0b6a5e1e4e8fioouiouqiuioasaz"
+user_id = 999041
+
+
+def ask(question, user_id):
+    # 设置请求头
+    headers = {
+        'token': token,
+        'X-API-KEY': api_key,
+        'Content-Type': 'application/json'
+    }
+
+    # 构造请求体
+    payload = json.dumps({
+        "msg": question,
+        # 历史问答，没有则为空List
+        "history": [],
+        "user_id": user_id,
+        "model": "deepseek-r1",      # 默认值 不用改
+        "using_indicator": True,     # 是否用指标
+        "start_time": "2024-01-01",  # 开始时间
+        "doc_show_type": ["A001", "A002", "A003", "A004"],  # 文档类型
+        "simple_tracking": simple_tracking  # 是否简单溯源
+    })
+    print(f"******开始提问:[{question}]")
+
+    # 发送请求
+    response = requests.request("POST", qa_url, data=payload, headers=headers, stream=True)
+
+    qa_result = ''
+
+    # 判断请求是否成功
+    if response.status_code == 200:
+        if stream_enabled:
+            try:
+                for chunk in response.iter_content(chunk_size=128):
+                    try:
+                        chunk_event = chunk.decode('utf-8', 'ignore')
+                    except UnicodeDecodeError as e:
+                        # 自定义处理解码错误，例如替换无法解码的部分
+                        chunk_event = chunk.decode('utf-8', 'replace')
+                        print(f"Decoding error occurred: {e}")
+                    qa_result += chunk_event
+                    print(f"\033[1;32m" + chunk_event)
+            except ChunkedEncodingError:
+                print("Stream ended prematurely. Handling gracefully.")
+
+        else:
+            # 获取响应内容
+            qa_result = response.content
+            # 将响应内容解码为utf-8格式
+            qa_result = qa_result.decode('utf-8')
+    else:
+        print(f"Failed to get stream data. Status code: {response.status_code}")
+        # 返回结果
+
+    return qa_result
+
+
+if __name__ == '__main__':
+    # 问题内容
+    question = '科大讯飞业绩怎么样?'
+    # 关闭吐字模式
+    stream_enabled = True
+    # 开启简单溯源
+    simple_tracking = True
+    # 调用函数进行问答
+    result = ask(question, user_id)
+
+    # 仅打印最终问答结果
+    print("**************COT**************")
+    cot_list = re.findall(r'\{"id":"_cot","content":"(.*?)"}', result)
+    cot = "".join(cot_list)
+    print(cot)
+    print("**********************************")
+
+    # 仅打印最终问答结果
+    print("**************最终答案**************")
+    print(re.findall(r'\{"id":"_final","content":"(.*?)"}', result)[0])
+    # print(result['answer'])
+    print("**********************************")
+
+    if simple_tracking:
+        print("**************溯源文件**************")
+        source_file = re.findall(r'\{"id":"tracking_documents","content":\s*(\[[^]]*])}', result)
+        if source_file and source_file.__len__() > 0:
+            print(source_file[0])
+        print("**********************************")

docs/user-guide.md

@@ -275,3 +275,4 @@ A:
 
 
 
+

frontend/src/app/config/page.tsx

@@ -27,13 +27,20 @@ export default function ConfigPage() {
   const [dbUrl, setDbUrl] = useState('');
   const [newApiApiKey, setNewApiApiKey] = useState('');
   const [newApiBaseUrl, setNewApiBaseUrl] = useState('');
+  const [alphaEngineApiUrl, setAlphaEngineApiUrl] = useState('');
+  const [alphaEngineApiKey, setAlphaEngineApiKey] = useState('');
+  const [alphaEngineToken, setAlphaEngineToken] = useState('');
+  const [alphaEngineUserId, setAlphaEngineUserId] = useState('');
   const [tushareApiKey, setTushareApiKey] = useState('');
   const [finnhubApiKey, setFinnhubApiKey] = useState('');
   
-  // 分析配置的本地状态
+  // 全局 LLM 配置
+  const [llmProvider, setLlmProvider] = useState('');
+  const [llmModel, setLlmModel] = useState('');
+  
+  // 分析配置的本地状态（移除 provider 和 model）
   const [localAnalysisConfig, setLocalAnalysisConfig] = useState<Record<string, {
     name: string;
-    model: string;
     prompt_template: string;
     dependencies?: string[];
   }>>({});
@@ -52,12 +59,40 @@ export default function ConfigPage() {
   // 初始化分析配置的本地状态
   useEffect(() => {
     if (analysisConfig?.analysis_modules) {
-      setLocalAnalysisConfig(analysisConfig.analysis_modules);
+      // 移除每个模块的 provider 和 model 字段
+      const cleanedConfig: typeof localAnalysisConfig = {};
+      Object.entries(analysisConfig.analysis_modules).forEach(([key, value]: [string, any]) => {
+        cleanedConfig[key] = {
+          name: value.name || '',
+          prompt_template: value.prompt_template || '',
+          dependencies: value.dependencies || []
+        };
+      });
+      setLocalAnalysisConfig(cleanedConfig);
     }
   }, [analysisConfig]);
   
-  // 更新分析配置中的某个字段
-  const updateAnalysisField = (type: string, field: 'name' | 'model' | 'prompt_template', value: string) => {
+  // 初始化全局 LLM 配置（从后端获取）
+  useEffect(() => {
+    const loadLlmConfig = async () => {
+      try {
+        const response = await fetch('/api/config/llm');
+        if (response.ok) {
+          const data = await response.json();
+          setLlmProvider(data.provider || '');
+          // 从 provider 配置中获取 model
+          const providerConfig = data.config || {};
+          setLlmModel(providerConfig.model || '');
+        }
+      } catch (e) {
+        console.error('Failed to load LLM config:', e);
+      }
+    };
+    loadLlmConfig();
+  }, []);
+  
+  // 更新分析配置中的某个字段（移除 provider 和 model）
+  const updateAnalysisField = (type: string, field: 'name' | 'prompt_template', value: string) => {
     setLocalAnalysisConfig(prev => ({
       ...prev,
       [type]: {
@@ -161,6 +196,15 @@ export default function ConfigPage() {
       };
     }
 
+    if (alphaEngineApiUrl || alphaEngineApiKey || alphaEngineToken) {
+      newConfig.alpha_engine = {
+        api_url: alphaEngineApiUrl || config?.alpha_engine?.api_url || '',
+        api_key: alphaEngineApiKey || config?.alpha_engine?.api_key || '',
+        token: alphaEngineToken || config?.alpha_engine?.token || '',
+        user_id: alphaEngineUserId ? parseInt(alphaEngineUserId) : (config?.alpha_engine?.user_id || 999041),
+      };
+    }
+
     if (tushareApiKey || finnhubApiKey) {
       newConfig.data_sources = {
         ...config?.data_sources,
@@ -216,10 +260,23 @@ export default function ConfigPage() {
     handleTest('finnhub', { api_key: finnhubApiKey || config?.data_sources?.finnhub?.api_key });
   };
 
+  const handleTestAlphaEngine = () => {
+    handleTest('alpha_engine', { 
+      api_url: alphaEngineApiUrl || config?.alpha_engine?.api_url,
+      api_key: alphaEngineApiKey || config?.alpha_engine?.api_key,
+      token: alphaEngineToken || config?.alpha_engine?.token,
+      user_id: alphaEngineUserId ? parseInt(alphaEngineUserId) : (config?.alpha_engine?.user_id || 999041)
+    });
+  };
+
   const handleReset = () => {
     setDbUrl('');
     setNewApiApiKey('');
     setNewApiBaseUrl('');
+    setAlphaEngineApiUrl('');
+    setAlphaEngineApiKey('');
+    setAlphaEngineToken('');
+    setAlphaEngineUserId('');
     setTushareApiKey('');
     setFinnhubApiKey('');
     setTestResults({});
@@ -345,10 +402,91 @@ export default function ConfigPage() {
           <Card>
             <CardHeader>
               <CardTitle>AI 服务配置</CardTitle>
-              <CardDescription>New API 设置 (兼容 OpenAI 格式)</CardDescription>
+              <CardDescription>配置大模型服务提供商和全局设置</CardDescription>
             </CardHeader>
-            <CardContent className="space-y-4">
+            <CardContent className="space-y-6">
+              {/* 全局 LLM 配置 */}
+              <div className="space-y-4 p-4 bg-muted/50 rounded-lg">
+                <h3 className="font-semibold">全局大模型设置</h3>
+                <p className="text-sm text-muted-foreground">
+                  所有分析模块将统一使用以下配置
+                </p>
+                <div className="space-y-4">
                   <div className="space-y-2">
+                    <Label htmlFor="llm-provider">大模型提供商</Label>
+                    <select
+                      id="llm-provider"
+                      value={llmProvider}
+                      onChange={(e) => setLlmProvider(e.target.value)}
+                      className="flex h-10 w-full rounded-md border border-input bg-background px-3 py-2 text-sm ring-offset-background file:border-0 file:bg-transparent file:text-sm file:font-medium placeholder:text-muted-foreground focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring focus-visible:ring-offset-2 disabled:cursor-not-allowed disabled:opacity-50"
+                    >
+                      <option value="">请选择提供商</option>
+                      <option value="new_api">New API (OpenAI 兼容)</option>
+                      <option value="gemini">Gemini</option>
+                      <option value="alpha_engine">AlphaEngine</option>
+                    </select>
+                  </div>
+                  
+                  <div className="space-y-2">
+                    <Label htmlFor="llm-model">模型名称</Label>
+                    <Input
+                      id="llm-model"
+                      type="text"
+                      value={llmModel}
+                      onChange={(e) => setLlmModel(e.target.value)}
+                      placeholder="例如: gemini-1.5-pro, deepseek-r1"
+                    />
+                    <p className="text-xs text-muted-foreground">
+                      根据选择的提供商输入对应的模型名称
+                    </p>
+                  </div>
+                  
+                  <Button 
+                    onClick={async () => {
+                      if (!llmProvider) {
+                        setSaveMessage('请先选择大模型提供商');
+                        return;
+                      }
+                      try {
+                        const response = await fetch('/api/config/llm', {
+                          method: 'PUT',
+                          headers: { 'Content-Type': 'application/json' },
+                          body: JSON.stringify({
+                            provider: llmProvider,
+                            model: llmModel || undefined
+                          })
+                        });
+                        if (response.ok) {
+                          setSaveMessage('全局 LLM 配置保存成功！');
+                        } else {
+                          const data = await response.json();
+                          setSaveMessage(`保存失败: ${data.detail || '未知错误'}`);
+                        }
+                      } catch (e: any) {
+                        setSaveMessage(`保存失败: ${e.message}`);
+                      }
+                      setTimeout(() => setSaveMessage(''), 5000);
+                    }}
+                    variant="outline"
+                  >
+                    保存全局 LLM 配置
+                  </Button>
+                  {saveMessage && saveMessage.includes('LLM') && (
+                    <Badge variant={saveMessage.includes('成功') ? 'default' : 'destructive'}>
+                      {saveMessage}
+                    </Badge>
+                  )}
+                </div>
+              </div>
+
+              <Separator />
+
+              <div className="space-y-4">
+                <div>
+                  <Label className="text-base font-medium">New API (OpenAI 兼容)</Label>
+                  <p className="text-sm text-muted-foreground mb-2">兼容 OpenAI API 格式的服务</p>
+                  <div className="space-y-2">
+                    <div>
                       <Label htmlFor="new-api-key">API Key</Label>
                       <div className="flex gap-2">
                         <Input
@@ -364,13 +502,13 @@ export default function ConfigPage() {
                         </Button>
                       </div>
                       {testResults.new_api && (
-                  <Badge variant={testResults.new_api.success ? 'default' : 'destructive'}>
+                        <Badge variant={testResults.new_api.success ? 'default' : 'destructive'} className="mt-2">
                           {testResults.new_api.message}
                         </Badge>
                       )}
                     </div>
                     
-              <div className="space-y-2">
+                    <div>
                       <Label htmlFor="new-api-base-url">Base URL</Label>
                       <Input
                         id="new-api-base-url"
@@ -381,6 +519,72 @@ export default function ConfigPage() {
                         className="flex-1"
                       />
                     </div>
+                  </div>
+                </div>
+
+                <Separator />
+
+                <div>
+                  <Label className="text-base font-medium">AlphaEngine</Label>
+                  <p className="text-sm text-muted-foreground mb-2">熵简科技投研问答 API</p>
+                  <div className="space-y-2">
+                    <div>
+                      <Label htmlFor="alpha-engine-api-url">API URL</Label>
+                      <Input
+                        id="alpha-engine-api-url"
+                        type="text"
+                        value={alphaEngineApiUrl}
+                        onChange={(e) => setAlphaEngineApiUrl(e.target.value)}
+                        placeholder="例如: http://api-ai-prod.valuesimplex.tech"
+                        className="flex-1"
+                      />
+                    </div>
+                    <div>
+                      <Label htmlFor="alpha-engine-api-key">API Key</Label>
+                      <div className="flex gap-2">
+                        <Input
+                          id="alpha-engine-api-key"
+                          type="password"
+                          value={alphaEngineApiKey}
+                          onChange={(e) => setAlphaEngineApiKey(e.target.value)}
+                          placeholder="留空表示保持当前值"
+                          className="flex-1"
+                        />
+                        <Button onClick={handleTestAlphaEngine} variant="outline">
+                          测试
+                        </Button>
+                      </div>
+                      {testResults.alpha_engine && (
+                        <Badge variant={testResults.alpha_engine.success ? 'default' : 'destructive'} className="mt-2">
+                          {testResults.alpha_engine.message}
+                        </Badge>
+                      )}
+                    </div>
+                    <div>
+                      <Label htmlFor="alpha-engine-token">Token</Label>
+                      <Input
+                        id="alpha-engine-token"
+                        type="password"
+                        value={alphaEngineToken}
+                        onChange={(e) => setAlphaEngineToken(e.target.value)}
+                        placeholder="留空表示保持当前值"
+                        className="flex-1"
+                      />
+                    </div>
+                    <div>
+                      <Label htmlFor="alpha-engine-user-id">User ID</Label>
+                      <Input
+                        id="alpha-engine-user-id"
+                        type="number"
+                        value={alphaEngineUserId}
+                        onChange={(e) => setAlphaEngineUserId(e.target.value)}
+                        placeholder="默认: 999041"
+                        className="flex-1"
+                      />
+                    </div>
+                  </div>
+                </div>
+              </div>
             </CardContent>
           </Card>
         </TabsContent>
@@ -470,19 +674,6 @@ export default function ConfigPage() {
                       />
                     </div>
 
-                    <div className="space-y-2">
-                      <Label htmlFor={`${type}-model`}>模型名称</Label>
-                      <Input
-                        id={`${type}-model`}
-                        value={config.model || ''}
-                        onChange={(e) => updateAnalysisField(type, 'model', e.target.value)}
-                        placeholder="例如: gemini-1.5-pro"
-                      />
-                      <p className="text-xs text-muted-foreground">
-                        在 AI 服务中配置的模型名称
-                      </p>
-                    </div>
-
                     <div className="space-y-2">
                       <Label>模块依赖</Label>
                       <div className="grid grid-cols-2 sm:grid-cols-3 md:grid-cols-4 gap-2 rounded-lg border p-4">

frontend/src/app/fonts/README.md

@@ -21,3 +21,4 @@
 
 
 
+

frontend/src/app/report/[symbol]/page.tsx

@@ -110,6 +110,9 @@ export default function ReportPage() {
   const [saving, setSaving] = useState(false)
   const [saveMsg, setSaveMsg] = useState<string | null>(null)
   
+  // TradingView 显示控制
+  const [showTradingView, setShowTradingView] = useState(false)
+
   const saveReport = async () => {
     try {
       setSaving(true)
@@ -155,6 +158,9 @@ export default function ReportPage() {
       return;
     }
 
+    // 标记已触发分析
+    fullAnalysisTriggeredRef.current = true;
+
     // 初始化/重置状态，准备顺序执行
     stopRequestedRef.current = false;
     abortControllerRef.current?.abort();
@@ -182,12 +188,7 @@ export default function ReportPage() {
     setManualRunKey((k) => k + 1);
   };
 
-  useEffect(() => {
-    if (financials && !fullAnalysisTriggeredRef.current) {
-      fullAnalysisTriggeredRef.current = true;
-      runFullAnalysis();
-    }
-  }, [financials]);
+  // 移除自动开始分析的逻辑，改为手动触发
 
   // 计算完成比例
   const completionProgress = useMemo(() => {
@@ -796,23 +797,53 @@ export default function ReportPage() {
           </TabsList>
           
           <TabsContent value="chart" className="space-y-4">
+            <div className="flex items-center justify-between mb-4">
               <h2 className="text-lg font-medium">股价图表（来自 TradingView）</h2>
+              <Button 
+                onClick={() => setShowTradingView(!showTradingView)}
+                variant={showTradingView ? "outline" : "default"}
+              >
+                {showTradingView ? '隐藏图表' : '显示图表'}
+              </Button>
+            </div>
+            
+            {showTradingView ? (
+              <>
                 <div className="flex items-center gap-3 text-sm mb-4">
                   <CheckCircle className="size-4 text-green-600" />
                   <div className="text-muted-foreground">
                     实时股价图表 - {unifiedSymbol}
                   </div>
                 </div>
-            
                 <TradingViewWidget 
                   symbol={unifiedSymbol} 
                   market={marketParam}
                   height={500}
                 />
+              </>
+            ) : (
+              <div className="flex items-center justify-center p-8 border border-dashed rounded-lg">
+                <div className="text-center space-y-2">
+                  <p className="text-muted-foreground">点击"显示图表"按钮加载 TradingView 股价图表</p>
+                  <p className="text-xs text-muted-foreground">图表数据来自 TradingView</p>
+                </div>
+              </div>
+            )}
           </TabsContent>
           
           <TabsContent value="financial" className="space-y-4">
+            <div className="flex items-center justify-between">
               <h2 className="text-lg font-medium">财务数据</h2>
+              {financials && !fullAnalysisTriggeredRef.current && analysisConfig?.analysis_modules && (
+                <Button 
+                  onClick={runFullAnalysis}
+                  disabled={isAnalysisRunningRef.current}
+                  className="ml-auto"
+                >
+                  开始分析
+                </Button>
+              )}
+            </div>
             <div className="flex items-center gap-3 text-sm">
               {isLoading ? (
                 <Spinner className="size-4" />
@@ -834,6 +865,25 @@ export default function ReportPage() {
               </div>
             )}
 
+            {financials && !fullAnalysisTriggeredRef.current && analysisConfig?.analysis_modules && (
+              <div className="bg-blue-50 border border-blue-200 rounded-lg p-4">
+                <div className="flex items-start gap-3">
+                  <div className="flex-1">
+                    <h3 className="font-medium text-blue-900 mb-1">准备开始分析</h3>
+                    <p className="text-sm text-blue-700">
+                      财务数据已加载完成。点击"开始分析"按钮启动大模型分析流程。
+                    </p>
+                  </div>
+                  <Button 
+                    onClick={runFullAnalysis}
+                    disabled={isAnalysisRunningRef.current}
+                  >
+                    开始分析
+                  </Button>
+                </div>
+              </div>
+            )}
+
 
 
             {financials && (

frontend/src/lib/prisma.ts

@@ -43,3 +43,4 @@ if (process.env.NODE_ENV !== 'production') globalForPrisma.prisma = prisma
 
 
 
+