feat: 增加AI分析并调通

app/analysis_db.py

@@ -33,4 +33,8 @@ def get_analysis_db():
 
 def init_analysis_db():
     """初始化期货智析数据库表"""
+    # 确保导入所有模型类，使其注册到 AnalysisBase
+    from app import analysis_models
+    # 直接导入 analysis_models 模块中的所有类
+    from app.analysis_models import FuturesAnalysis, WatchedSymbol, AIModelConfig, AnalysisSettings, AIAnalysisCache
     AnalysisBase.metadata.create_all(bind=analysis_engine)

app/analysis_models.py

@@ -86,3 +86,16 @@ class AnalysisSettings(AnalysisBase):
 
     def __repr__(self):
         return f"<AnalysisSettings {self.key}>"
+
+
+class AIAnalysisCache(AnalysisBase):
+    """AI分析缓存表"""
+    __tablename__ = "ai_analysis_cache"
+
+    id = Column(Integer, primary_key=True, autoincrement=True)
+    symbol = Column(String(32), nullable=False, index=True, comment="品种合约代码")
+    analysis_data = Column(JSON, nullable=False, comment="AI分析结果数据")
+    created_at = Column(DateTime, nullable=False, default=datetime.now, index=True, comment="分析时间")
+
+    def __repr__(self):
+        return f"<AIAnalysisCache {self.symbol} {self.created_at}>"

app/api/futures_analysis.py

@@ -12,7 +12,7 @@ from sqlalchemy.orm import Session
 
 from app.database import get_db
 from app.analysis_db import get_analysis_db
-from app.analysis_models import FuturesAnalysis, WatchedSymbol, AIModelConfig, AnalysisSettings
+from app.analysis_models import FuturesAnalysis, WatchedSymbol, AIModelConfig, AnalysisSettings, AIAnalysisCache
 from app.services.cache import get_cached_data, get_latest_cached, save_market_data
 from app.services.collector import fetch_symbol_data
 
@@ -694,6 +694,7 @@ def delete_ai_model(model_id: int, adb: Session = Depends(get_analysis_db)):
 # ==================== 数据刷新接口 ====================
 
 from app.services.cache import needs_refresh, get_symbol_timestamp
+from app.services.ai_analysis import AIFuturesAnalyzer
 
 refresh_lock = threading.Lock()
 refresh_status = {"running": False, "progress": 0, "total": 0, "message": ""}
@@ -786,3 +787,112 @@ def refresh_all_symbols_api(background_tasks: BackgroundTasks):
 def get_refresh_status():
     """获取刷新状态"""
     return {"success": True, "data": refresh_status}
+
+
+# ==================== AI智能分析接口 ====================
+
+@router.post("/ai-analysis/{symbol}")
+def run_ai_analysis(symbol: str, db: Session = Depends(get_db), analysis_db: Session = Depends(get_analysis_db)):
+    """执行AI智能分析"""
+    try:
+        analyzer = AIFuturesAnalyzer(db, analysis_db)
+        result = analyzer.analyze(symbol)
+        
+        if result.get("success"):
+            return {
+                "success": True,
+                "data": result["data"]
+            }
+        else:
+            return {
+                "success": False,
+                "error": result.get("error", "AI分析失败")
+            }
+    except Exception as e:
+        logger.error(f"AI分析失败: {e}")
+        return {
+            "success": False,
+            "error": f"AI分析失败: {str(e)}"
+        }
+
+
+@router.get("/ai-analysis/{symbol}")
+def get_ai_analysis(symbol: str, force_refresh: bool = False, db: Session = Depends(get_db), analysis_db: Session = Depends(get_analysis_db)):
+    """获取AI分析结果（可选择是否强制刷新）"""
+    try:
+        analyzer = AIFuturesAnalyzer(db, analysis_db)
+        
+        if force_refresh:
+            result = analyzer.analyze(symbol)
+            if result.get("success"):
+                return {
+                    "success": True,
+                    "data": result["data"],
+                    "is_cached": False
+                }
+            else:
+                return {
+                    "success": False,
+                    "error": result.get("error", "AI分析失败")
+                }
+        
+        cache = analyzer.get_latest_cache(symbol)
+        if cache:
+            return {
+                "success": True,
+                "data": {
+                    "id": cache.id,
+                    "symbol": cache.symbol,
+                    "analysis_time": cache.created_at.isoformat(),
+                    "result": cache.analysis_data
+                },
+                "is_cached": True
+            }
+        
+        result = analyzer.analyze(symbol)
+        if result.get("success"):
+            return {
+                "success": True,
+                "data": result["data"],
+                "is_cached": False
+            }
+        else:
+            return {
+                "success": False,
+                "error": result.get("error", "未找到分析结果")
+            }
+    except Exception as e:
+        logger.error(f"获取AI分析结果失败: {e}")
+        return {
+            "success": False,
+            "error": f"获取AI分析失败: {str(e)}"
+        }
+
+
+@router.get("/ai-analysis/{symbol}/history")
+def get_ai_analysis_history(symbol: str, limit: int = 10, analysis_db: Session = Depends(get_analysis_db)):
+    """获取AI分析历史记录"""
+    try:
+        records = analysis_db.query(AIAnalysisCache).filter(
+            AIAnalysisCache.symbol == symbol
+        ).order_by(
+            AIAnalysisCache.created_at.desc()
+        ).limit(limit).all()
+        
+        return {
+            "success": True,
+            "data": [{
+                "id": r.id,
+                "symbol": r.symbol,
+                "analysis_time": r.created_at.isoformat(),
+                "summary": r.analysis_data.get("summary", ""),
+                "trading_suggestion": r.analysis_data.get("trading_suggestion", {}),
+                "confidence": r.analysis_data.get("trading_suggestion", {}).get("confidence", 0)
+            } for r in records]
+        }
+    except Exception as e:
+        logger.error(f"获取AI分析历史失败: {e}")
+        return {
+            "success": False,
+            "error": f"获取历史记录失败: {str(e)}"
+        }

app/services/ai_analysis.py

@@ -0,0 +1,459 @@
+"""
+AI分析服务 - 期货四维联合分析
+"""
+import json
+import re
+import logging
+from datetime import datetime
+from typing import Dict, List, Optional
+from sqlalchemy.orm import Session
+
+from app.analysis_models import AIAnalysisCache
+from app.services.cache import get_cached_data, get_latest_cached
+from pathlib import Path
+
+CONFIG_DIR = Path(__file__).resolve().parent.parent.parent / "config"
+AI_CONFIG_FILE = CONFIG_DIR / "ai_config.json"
+
+logger = logging.getLogger(__name__)
+
+
+class AIAnalysisPrompt:
+    """AI分析提示词管理器"""
+    
+    SYSTEM_PROMPT = """你是一位拥有20年实战经验的资深金融交易分析师，精通A股市场与商品期货的技术分析。
+你的核心使命是基于提供的K线数据，执行【四维联合判断分析法(4D-XV)】，并提供包含风控红线审查的客观交易策略。
+
+你的分析必须遵循以下原则：
+1. 数据驱动：所有结论必须基于输入的JSON数据，严禁凭空捏造。
+2. 四维共振：任何交易建议必须经过MACD（趋势）、成交量（资金）、KDJ（时机）、多周期（方向）的交叉验证。
+3. 红线否决：如果数据触发【17条交易红线】，必须直接给出【禁止交易】或【止损】的建议。
+4. 客观中立：不使用绝对化表述，提供情景预案（概率估算）。
+5. 动态切换：遇到关键位放量突破时，立即切换右侧思维，不逆势死扛。"""
+
+    ANALYSIS_TEMPLATE = """请严格按照以下JSON格式输出分析结果（不要输出任何其他内容）：
+
+{
+  "summary": "一句话总结当前市场状态",
+  "four_dimensional": {
+    "60min": {
+      "macd": {"trend": "up/down/neutral", "histogram": "放大/缩小/背离", "position": "零轴上/下"},
+      "volume": {"status": "放量上涨/缩量回调/趋势量能/拐点量能", "ratio": 1.5},
+      "kdj": {"k": 85, "d": 80, "j": 95, "status": "超买/超卖/中性", "signal": "金叉/死叉/钝化"},
+      "conclusion": "定大势结论"
+    },
+    "30min": {
+      "macd": {"trend": "up/down/neutral", "histogram": "放大/缩小/背离", "position": "零轴上/下"},
+      "volume": {"status": "放量/缩量/正常", "ratio": 1.5},
+      "kdj": {"k": 50, "d": 45, "j": 60, "status": "中性", "signal": "金叉/死叉"},
+      "conclusion": "找拐点结论"
+    },
+    "15min": {
+      "macd": {"trend": "up/down/neutral", "histogram": "放大/缩小/背离", "position": "零轴上/下"},
+      "volume": {"status": "放量/缩量/正常", "ratio": 2.0},
+      "kdj": {"k": 30, "d": 25, "j": 40, "status": "超卖", "signal": "金叉/死叉"},
+      "conclusion": "择入场结论"
+    }
+  },
+  "kdj_diagnosis": {
+    "current_status": "超买/超卖/中性区域",
+    "divergence": "是否存在顶/底背离",
+    "paralysis": "是否钝化（持续>6根K线）",
+    "recommendation": "KDJ使用建议"
+  },
+  "pivot_points": {
+    "r2": 7500,
+    "r1": 7350,
+    "pp": 7200,
+    "s1": 7050,
+    "s2": 6900,
+    "validation": {
+      "test_count": 3,
+      "volume_confirmed": true,
+      "multi_period_resonance": true,
+      "breakback_confirmed": false
+    }
+  },
+  "red_lines_check": {
+    "passed": true,
+    "violated": [],
+    "warnings": ["暂无红线警告"]
+  },
+  "discipline_score": {
+    "total": 9,
+    "max": 11,
+    "details": {
+      "trend": true,
+      "position": true,
+      "signal": true,
+      "risk": true,
+      "mindset": true
+    }
+  },
+  "trading_suggestion": {
+    "direction": "做多/做空/观望",
+    "confidence": 75,
+    "entry_range": {"min": 7050, "max": 7100},
+    "stop_loss": 6950,
+    "take_profit": [{"price": 7200, "ratio": 50}, {"price": 7350, "ratio": 30}, {"price": 7500, "ratio": 20}],
+    "position_size": "轻仓/半仓/重仓",
+    "reason": "做多理由"
+  },
+  "scenario_plans": {
+    "breakthrough": {"probability": 35, "action": "放量突破关键位，跟随右侧思维"},
+    "consolidation": {"probability": 40, "action": "R1-S1区间内高抛低吸"},
+    "reversal": {"probability": 15, "action": "MACD顶/底背离+量能不足，立即止损反手"},
+    "news_impact": {"probability": 10, "action": "减仓50%规避不确定性"}
+  },
+  "risk_warnings": [
+    "技术指标具有滞后性，历史表现不代表未来",
+    "需结合基本面和市场情绪综合判断"
+  ],
+  "experience_lessons": [
+    "警惕缩量创新高，可能是诱多信号",
+    "KDJ超买钝化中不宜逆势做空"
+  ]
+}"""
+
+    @classmethod
+    def build_prompt(cls, symbol: str, data: Dict) -> str:
+        """构建完整的AI分析提示词"""
+        prompt = f"""{cls.SYSTEM_PROMPT}
+
+现在请分析以下期货品种的K线数据：
+
+## 品种信息
+- 合约代码：{symbol}
+- 当前价格：{data.get('current_price', 'N/A')}
+
+## 多周期K线数据
+```json
+{json.dumps(data, ensure_ascii=False, indent=2)}
+```
+
+{cls.ANALYSIS_TEMPLATE}"""
+
+        return prompt
+
+
+class AIFuturesAnalyzer:
+    """AI期货分析器"""
+    
+    def __init__(self, db: Session, analysis_db: Session = None):
+        self.db = db
+        self.analysis_db = analysis_db or db
+    
+    def get_active_model(self) -> Optional[Dict]:
+        """获取当前激活的AI模型配置"""
+        try:
+            if not AI_CONFIG_FILE.exists():
+                return None
+            
+            with open(AI_CONFIG_FILE, "r", encoding="utf-8") as f:
+                config = json.load(f)
+            
+            models = config.get("models", [])
+            active_model_name = config.get("active_model")
+            
+            if active_model_name:
+                for model in models:
+                    if model.get("model_name") == active_model_name and model.get("enabled", True):
+                        return model
+            
+            for model in models:
+                if model.get("enabled", True):
+                    logger.warning(f"未找到匹配的激活模型，使用第一个启用的模型: {model.get('model_name')}")
+                    return model
+            
+            return None
+        except Exception as e:
+            logger.error(f"加载AI配置失败: {e}")
+            return None
+    
+    def prepare_multi_period_data(self, symbol: str) -> Optional[Dict]:
+        """准备多周期数据用于AI分析"""
+        cached_data = get_cached_data(
+            self.db, 
+            symbol, 
+            "futures",
+            ["5min", "15min", "30min", "60min", "daily"]
+        )
+        
+        if not cached_data or not cached_data.get("timeframes"):
+            return None
+        
+        timeframes = cached_data.get("timeframes", {})
+        current_price = cached_data.get("current_price")
+        
+        result = {
+            "symbol": symbol,
+            "current_price": current_price,
+            "timeframes": {}
+        }
+        
+        for period_name, db_period in [("5min", "5min"), ("15min", "15min"), 
+                                        ("30min", "30min"), ("60min", "60min"), 
+                                        ("daily", "daily")]:
+            if period_name in timeframes and timeframes[period_name]:
+                candles = timeframes[period_name]
+                if len(candles) >= 20:
+                    result["timeframes"][period_name] = self._analyze_timeframe(candles, period_name)
+        
+        return result
+    
+    def _analyze_timeframe(self, candles: List[Dict], period: str) -> Dict:
+        """分析单个周期的技术指标"""
+        if not candles or len(candles) < 20:
+            return {}
+        
+        closes = [float(c.get("close", 0)) for c in candles]
+        highs = [float(c.get("high", 0)) for c in candles]
+        lows = [float(c.get("low", 0)) for c in candles]
+        volumes = [float(c.get("volume", 0)) for c in candles]
+        
+        ma10 = sum(closes[-10:]) / 10 if len(closes) >= 10 else None
+        ma20 = sum(closes[-20:]) / 20 if len(closes) >= 20 else None
+        
+        macd_data = self._calc_macd(closes)
+        kdj_data = self._calc_kdj(highs, lows, closes)
+        
+        avg_volume = sum(volumes[-20:]) / 20 if len(volumes) >= 20 else 0
+        current_volume = volumes[-1] if volumes else 0
+        
+        return {
+            "trend": "up" if closes[-1] > closes[0] else "down",
+            "ma10": round(ma10, 2) if ma10 else None,
+            "ma20": round(ma20, 2) if ma20 else None,
+            "macd_dif": round(macd_data["dif"], 4),
+            "macd_dea": round(macd_data["dea"], 4),
+            "macd_histogram": round(macd_data["histogram"], 4),
+            "kdj_k": kdj_data["k"],
+            "kdj_d": kdj_data["d"],
+            "kdj_j": kdj_data["j"],
+            "volume_avg": round(avg_volume, 2),
+            "volume_current": round(current_volume, 2),
+            "volume_ratio": round(current_volume / avg_volume, 2) if avg_volume > 0 else 1,
+            "candles": candles[-10:] if len(candles) > 10 else candles
+        }
+    
+    def _calc_macd(self, closes: List[float]) -> Dict:
+        """计算MACD指标"""
+        if len(closes) < 26:
+            return {"dif": 0, "dea": 0, "histogram": 0}
+        
+        ema12 = self._calc_ema(closes, 12)
+        ema26 = self._calc_ema(closes, 26)
+        
+        dif = ema12 - ema26
+        dea = self._calc_ema([dif] * len(closes), 9)
+        histogram = 2 * (dif - dea)
+        
+        return {"dif": dif, "dea": dea, "histogram": histogram}
+    
+    def _calc_ema(self, data: List[float], period: int) -> float:
+        """计算EMA"""
+        if len(data) < period:
+            return sum(data) / len(data) if data else 0
+        
+        multiplier = 2 / (period + 1)
+        ema = sum(data[:period]) / period
+        
+        for i in range(period, len(data)):
+            ema = (data[i] - ema) * multiplier + ema
+        
+        return ema
+    
+    def _calc_kdj(self, highs: List[float], lows: List[float], closes: List[float]) -> Dict:
+        """计算KDJ指标"""
+        if len(closes) < 9:
+            return {"k": 50, "d": 50, "j": 50}
+        
+        period = 9
+        recent_highs = highs[-period:]
+        recent_lows = lows[-period:]
+        recent_closes = closes[-period:]
+        
+        highest = max(recent_highs)
+        lowest = min(recent_lows)
+        current = recent_closes[-1]
+        
+        if highest == lowest:
+            rsv = 50
+        else:
+            rsv = (current - lowest) / (highest - lowest) * 100
+        
+        k = rsv * 2 / 3 + 50 / 3
+        d = k * 2 / 3 + 50 / 3
+        j = 3 * k - 2 * d
+        
+        return {"k": round(k, 2), "d": round(d, 2), "j": round(j, 2)}
+    
+    def call_ai_model(self, prompt: str, model: Dict) -> Optional[str]:
+        """调用AI模型"""
+        try:
+            import requests
+            
+            api_base = model.get("api_base", "https://api.openai.com/v1")
+            api_key = model.get("api_key", "")
+            model_id = model.get("model_id") or model.get("model_name", "")
+            
+            logger.info(f"========== AI模型调用开始 ==========")
+            logger.info(f"API Base: {api_base}")
+            logger.info(f"API Key: {'已配置' if api_key else '未配置'} ({api_key[:10]}...)" if api_key else "API Key: 未配置")
+            logger.info(f"Model ID: {model_id}")
+            logger.info(f"Temperature: {model.get('temperature', 0.7)}")
+            logger.info(f"Max Tokens: {model.get('max_tokens', 2000)}")
+            
+            if not api_key:
+                logger.error("API Key 未配置，无法调用AI模型")
+                return None
+            
+            headers = {
+                "Content-Type": "application/json",
+                "Authorization": f"Bearer {api_key}"
+            }
+            
+            payload = {
+                "model": model_id,
+                "messages": [
+                    {"role": "user", "content": prompt}
+                ],
+                "temperature": model.get("temperature", 0.7),
+                "max_tokens": model.get("max_tokens", 2000)
+            }
+            
+            url = f"{api_base}/chat/completions"
+            logger.info(f"请求 URL: {url}")
+            logger.info(f"请求 Payload 大小: {len(str(payload))} 字符")
+            
+            response = requests.post(
+                url,
+                headers=headers,
+                json=payload,
+                timeout=180  # 增加到180秒（3分钟）超时
+            )
+            
+            logger.info(f"响应状态码: {response.status_code}")
+            
+            if response.status_code == 200:
+                result = response.json()
+                content = result["choices"][0]["message"]["content"]
+                logger.info(f"AI响应成功，内容长度: {len(content)} 字符")
+                logger.info(f"========== AI模型调用成功 ==========")
+                return content
+            else:
+                logger.error(f"AI模型调用失败:")
+                logger.error(f"  状态码: {response.status_code}")
+                logger.error(f"  响应内容: {response.text}")
+                logger.error(f"========== AI模型调用失败 ==========")
+                return None
+                
+        except requests.exceptions.Timeout:
+            logger.error("AI模型调用超时（超过60秒）")
+            logger.error(f"========== AI模型调用失败 ==========")
+            return None
+        except requests.exceptions.ConnectionError as e:
+            logger.error(f"AI模型连接错误: {e}")
+            logger.error(f"========== AI模型调用失败 ==========")
+            return None
+        except requests.exceptions.RequestException as e:
+            logger.error(f"AI模型请求异常: {e}")
+            logger.error(f"========== AI模型调用失败 ==========")
+            return None
+        except Exception as e:
+            logger.error(f"调用AI模型未知异常: {e}", exc_info=True)
+            logger.error(f"========== AI模型调用失败 ==========")
+            return None
+    
+    def parse_ai_response(self, response: str) -> Optional[Dict]:
+        """解析AI返回的JSON响应"""
+        try:
+            json_match = re.search(r'\{[\s\S]*\}', response)
+            if json_match:
+                return json.loads(json_match.group(0))
+            return None
+        except Exception as e:
+            logger.error(f"解析AI响应失败: {e}")
+            return None
+    
+    def save_analysis_cache(self, symbol: str, analysis_data: Dict) -> AIAnalysisCache:
+        """保存AI分析结果到缓存"""
+        cache = AIAnalysisCache(
+            symbol=symbol,
+            analysis_data=analysis_data,
+            created_at=datetime.now()
+        )
+        
+        self.analysis_db.add(cache)
+        self.analysis_db.commit()
+        self.analysis_db.refresh(cache)
+        
+        return cache
+
+    def get_latest_cache(self, symbol: str) -> Optional[AIAnalysisCache]:
+        """获取最新的AI分析缓存"""
+        return self.analysis_db.query(AIAnalysisCache).filter(
+            AIAnalysisCache.symbol == symbol
+        ).order_by(AIAnalysisCache.created_at.desc()).first()
+    
+    def analyze(self, symbol: str) -> Dict:
+        """执行完整的AI分析流程"""
+        logger.info(f"===== 开始AI分析: {symbol} =====")
+        
+        model = self.get_active_model()
+        if not model:
+            logger.error("未找到激活的AI模型配置")
+            return {
+                "success": False,
+                "error": "未配置AI模型或模型未激活"
+            }
+        
+        logger.info(f"使用AI模型: {model.get('model_name')}")
+        
+        data = self.prepare_multi_period_data(symbol)
+        if not data:
+            logger.error(f"未找到 {symbol} 的市场数据")
+            return {
+                "success": False,
+                "error": f"未找到 {symbol} 的市场数据"
+            }
+        
+        logger.info(f"市场数据准备成功，包含周期: {list(data.get('timeframes', {}).keys())}")
+        
+        prompt = AIAnalysisPrompt.build_prompt(symbol, data)
+        logger.info(f"AI提示词生成完成，长度: {len(prompt)} 字符")
+        
+        response = self.call_ai_model(prompt, model)
+        
+        if not response:
+            logger.error("AI模型返回空响应")
+            return {
+                "success": False,
+                "error": "AI模型调用失败"
+            }
+        
+        logger.info(f"AI模型响应接收成功，长度: {len(response)} 字符")
+        
+        analysis_result = self.parse_ai_response(response)
+        if not analysis_result:
+            logger.error(f"AI响应解析失败，原始响应前100字符: {response[:100]}")
+            return {
+                "success": False,
+                "error": "AI响应解析失败"
+            }
+        
+        logger.info(f"AI响应解析成功")
+        
+        cache = self.save_analysis_cache(symbol, analysis_result)
+        logger.info(f"分析结果已保存到缓存，ID: {cache.id}")
+        logger.info(f"===== AI分析完成: {symbol} =====")
+        
+        return {
+            "success": True,
+            "data": {
+                "id": cache.id,
+                "symbol": symbol,
+                "analysis_time": cache.created_at.isoformat(),
+                "result": analysis_result
+            }
+        }

app/static/futures_analysis.css

@@ -1597,6 +1597,396 @@ body.theme-minimal .sort-select select:hover {
     border-color: var(--text-muted);
 }
 
+/* ============================================
+   AI智能分析样式
+   ============================================ */
+
+.panel-header-actions {
+    display: flex;
+    gap: 8px;
+    margin-left: auto;
+}
+
+.ai-analyze-btn {
+    display: flex;
+    align-items: center;
+    gap: 6px;
+    padding: 6px 12px;
+    background: linear-gradient(135deg, var(--purple), var(--cyan));
+    border: none;
+    border-radius: 6px;
+    color: white;
+    font-size: 12px;
+    font-weight: 500;
+    cursor: pointer;
+    transition: all 0.3s ease;
+}
+
+.ai-analyze-btn:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 4px 12px var(--purple-glow);
+}
+
+.ai-analyze-btn:active {
+    transform: translateY(0);
+}
+
+.ai-analyze-btn:disabled {
+    opacity: 0.6;
+    cursor: not-allowed;
+    transform: none;
+}
+
+.ai-analyze-btn i {
+    font-size: 11px;
+}
+
+.ai-analysis-content {
+    min-height: 120px;
+}
+
+.ai-analysis-placeholder {
+    display: flex;
+    flex-direction: column;
+    align-items: center;
+    justify-content: center;
+    padding: 24px;
+    color: var(--text-muted);
+}
+
+.ai-analysis-placeholder i {
+    font-size: 32px;
+    margin-bottom: 12px;
+    opacity: 0.5;
+}
+
+.ai-analysis-placeholder p {
+    font-size: 13px;
+    text-align: center;
+}
+
+.ai-analysis-loading {
+    display: flex;
+    flex-direction: column;
+    align-items: center;
+    justify-content: center;
+    padding: 24px;
+    gap: 12px;
+}
+
+.ai-analysis-loading i {
+    font-size: 28px;
+    color: var(--purple);
+    animation: pulse 1.5s ease-in-out infinite;
+}
+
+@keyframes pulse {
+    0%, 100% { opacity: 0.5; transform: scale(1); }
+    50% { opacity: 1; transform: scale(1.1); }
+}
+
+.ai-analysis-result {
+    display: flex;
+    flex-direction: column;
+    gap: 12px;
+}
+
+.ai-summary {
+    padding: 12px;
+    background: rgba(139, 92, 246, 0.08);
+    border: 1px solid rgba(139, 92, 246, 0.15);
+    border-radius: 8px;
+    font-size: 13px;
+    line-height: 1.6;
+    color: var(--text-primary);
+}
+
+.ai-suggestion-row {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    padding: 10px 12px;
+    background: rgba(255, 255, 255, 0.03);
+    border-radius: 6px;
+}
+
+.ai-suggestion-direction {
+    display: flex;
+    align-items: center;
+    gap: 8px;
+}
+
+.ai-suggestion-direction i {
+    font-size: 16px;
+}
+
+.ai-suggestion-direction.long i {
+    color: var(--green);
+}
+
+.ai-suggestion-direction.short i {
+    color: var(--red);
+}
+
+.ai-suggestion-direction.neutral i {
+    color: var(--amber);
+}
+
+.ai-confidence {
+    display: flex;
+    align-items: center;
+    gap: 6px;
+    font-size: 12px;
+    color: var(--text-secondary);
+}
+
+.ai-confidence-bar {
+    width: 60px;
+    height: 4px;
+    background: rgba(255, 255, 255, 0.1);
+    border-radius: 2px;
+    overflow: hidden;
+}
+
+.ai-confidence-fill {
+    height: 100%;
+    background: linear-gradient(90deg, var(--amber), var(--green));
+    border-radius: 2px;
+    transition: width 0.5s ease;
+}
+
+.ai-key-metrics {
+    display: grid;
+    grid-template-columns: repeat(2, 1fr);
+    gap: 8px;
+}
+
+.ai-metric-item {
+    display: flex;
+    justify-content: space-between;
+    padding: 8px 10px;
+    background: rgba(255, 255, 255, 0.02);
+    border-radius: 6px;
+    font-size: 12px;
+}
+
+.ai-metric-item .label {
+    color: var(--text-muted);
+}
+
+.ai-metric-item .value {
+    font-weight: 600;
+    color: var(--text-primary);
+}
+
+.ai-metric-item .value.up {
+    color: var(--green);
+}
+
+.ai-metric-item .value.down {
+    color: var(--red);
+}
+
+.ai-timestamp {
+    text-align: center;
+    font-size: 11px;
+    color: var(--text-muted);
+    padding-top: 8px;
+    border-top: 1px solid var(--border-color);
+}
+
+/* AI分析详情模态框 */
+.modal-large {
+    max-width: 900px;
+    width: 90%;
+}
+
+.modal-large .modal-body {
+    max-height: 80vh;
+    overflow-y: auto;
+}
+
+.ai-modal-section {
+    margin-bottom: 24px;
+}
+
+.ai-modal-section:last-child {
+    margin-bottom: 0;
+}
+
+.ai-modal-section-title {
+    display: flex;
+    align-items: center;
+    gap: 8px;
+    font-size: 15px;
+    font-weight: 600;
+    color: var(--cyan);
+    margin-bottom: 12px;
+    padding-bottom: 8px;
+    border-bottom: 1px solid var(--border-color);
+}
+
+.ai-modal-section-title i {
+    font-size: 16px;
+}
+
+.four-dimensional-table {
+    width: 100%;
+    border-collapse: collapse;
+    font-size: 13px;
+}
+
+.four-dimensional-table th,
+.four-dimensional-table td {
+    padding: 10px 12px;
+    text-align: left;
+    border-bottom: 1px solid var(--border-color);
+}
+
+.four-dimensional-table th {
+    background: rgba(139, 92, 246, 0.08);
+    color: var(--purple);
+    font-weight: 600;
+    font-size: 12px;
+}
+
+.four-dimensional-table td {
+    color: var(--text-secondary);
+}
+
+.four-dimensional-table tr:last-child td {
+    border-bottom: none;
+}
+
+.four-dimensional-table .period-cell {
+    font-weight: 600;
+    color: var(--text-primary);
+}
+
+.scenario-cards {
+    display: grid;
+    grid-template-columns: repeat(2, 1fr);
+    gap: 12px;
+}
+
+.scenario-card {
+    padding: 12px;
+    background: rgba(255, 255, 255, 0.03);
+    border-radius: 8px;
+    border-left: 3px solid var(--cyan);
+}
+
+.scenario-card.breakthrough {
+    border-left-color: var(--green);
+}
+
+.scenario-card.consolidation {
+    border-left-color: var(--amber);
+}
+
+.scenario-card.reversal {
+    border-left-color: var(--red);
+}
+
+.scenario-card.news {
+    border-left-color: var(--purple);
+}
+
+.scenario-probability {
+    display: inline-block;
+    padding: 2px 8px;
+    background: rgba(139, 92, 246, 0.15);
+    border-radius: 4px;
+    font-size: 11px;
+    font-weight: 600;
+    color: var(--purple);
+    margin-bottom: 6px;
+}
+
+.scenario-action {
+    font-size: 12px;
+    color: var(--text-secondary);
+    line-height: 1.5;
+}
+
+.red-lines-list {
+    display: flex;
+    flex-direction: column;
+    gap: 8px;
+}
+
+.red-line-item {
+    display: flex;
+    align-items: center;
+    gap: 8px;
+    padding: 8px 12px;
+    background: rgba(239, 68, 68, 0.08);
+    border: 1px solid rgba(239, 68, 68, 0.15);
+    border-radius: 6px;
+    font-size: 12px;
+    color: var(--red);
+}
+
+.red-line-item.pass {
+    background: rgba(16, 185, 129, 0.08);
+    border-color: rgba(16, 185, 129, 0.15);
+    color: var(--green);
+}
+
+.discipline-grid {
+    display: grid;
+    grid-template-columns: repeat(3, 1fr);
+    gap: 10px;
+}
+
+.discipline-item {
+    display: flex;
+    align-items: center;
+    gap: 8px;
+    padding: 10px;
+    background: rgba(255, 255, 255, 0.03);
+    border-radius: 6px;
+}
+
+.discipline-item i {
+    font-size: 16px;
+}
+
+.discipline-item.pass i {
+    color: var(--green);
+}
+
+.discipline-item.fail i {
+    color: var(--red);
+}
+
+.discipline-label {
+    font-size: 12px;
+    color: var(--text-secondary);
+}
+
+.ai-experience-list {
+    display: flex;
+    flex-direction: column;
+    gap: 8px;
+}
+
+.experience-item {
+    display: flex;
+    align-items: flex-start;
+    gap: 8px;
+    padding: 10px 12px;
+    background: rgba(245, 158, 11, 0.08);
+    border: 1px solid rgba(245, 158, 11, 0.15);
+    border-radius: 6px;
+    font-size: 12px;
+    color: var(--amber);
+}
+
+.experience-item i {
+    margin-top: 2px;
+}
+
 body.theme-minimal .sort-select select:focus {
     border-color: var(--cyan);
     box-shadow: 0 0 0 0.125rem rgba(73, 79, 223, 0.2);

app/static/futures_analysis.html

@@ -362,6 +362,26 @@
                             </div>
                         </div>
 
+                        <!-- AI智能分析 -->
+                        <div class="panel-card ai-analysis-card" id="ai-analysis-panel">
+                            <div class="panel-header">
+                                <i class="fas fa-brain"></i>
+                                <span>AI 四维分析</span>
+                                <div class="panel-header-actions">
+                                    <button class="ai-analyze-btn" id="ai-analyze-btn" onclick="runAIAnalysis()" title="执行AI分析">
+                                        <i class="fas fa-play"></i>
+                                        <span>智能分析</span>
+                                    </button>
+                                </div>
+                            </div>
+                            <div class="ai-analysis-content" id="ai-analysis-content">
+                                <div class="ai-analysis-placeholder">
+                                    <i class="fas fa-brain"></i>
+                                    <p>点击"智能分析"按钮获取AI分析结果</p>
+                                </div>
+                            </div>
+                        </div>
+
                         <!-- 趋势评分 -->
                         <div class="panel-card score-card">
                             <div class="panel-header">
@@ -385,6 +405,19 @@
         </main>
     </div>
 
+    <!-- AI分析详情对话框 -->
+    <div class="modal-overlay" id="ai-analysis-modal">
+        <div class="modal-content modal-large">
+            <div class="modal-header">
+                <h3><i class="fas fa-brain"></i> AI 四维联合分析报告</h3>
+                <button class="modal-close" onclick="closeModal('ai-analysis-modal')"><i class="fas fa-xmark"></i></button>
+            </div>
+            <div class="modal-body" id="ai-analysis-modal-body">
+                <!-- 动态生成 -->
+            </div>
+        </div>
+    </div>
+
     <!-- AI建议详情对话框 -->
     <div class="modal-overlay" id="suggestion-modal">
         <div class="modal-content">

app/static/futures_analysis.js

@@ -143,6 +143,7 @@ function showDetailView(symbol) {
     loadFuturesDetail(symbol);
     loadKlineData(symbol, currentPeriod);
     loadHistoryList(symbol);
+    loadAIAnalysis();
 }
 
 async function loadWatchedSymbols() {
@@ -530,6 +531,37 @@ async function loadHistoryList(symbol) {
     }
 }
 
+async function loadAIAnalysis() {
+    if (!currentSymbol) return;
+    
+    const content = document.getElementById('ai-analysis-content');
+    
+    try {
+        const response = await fetch(`${API_BASE}/ai-analysis/${currentSymbol}`);
+        const data = await response.json();
+        
+        if (data.success && data.data) {
+            currentAIAnalysis = data.data;
+            displayAIAnalysisResult(data.data);
+        } else {
+            content.innerHTML = `
+                <div class="ai-analysis-placeholder">
+                    <i class="fas fa-brain"></i>
+                    <p>点击"智能分析"按钮获取AI分析结果</p>
+                </div>
+            `;
+        }
+    } catch (error) {
+        console.error('加载AI分析失败:', error);
+        content.innerHTML = `
+            <div class="ai-analysis-placeholder">
+                <i class="fas fa-brain"></i>
+                <p>点击"智能分析"按钮获取AI分析结果</p>
+            </div>
+        `;
+    }
+}
+
 function renderHistoryList(records) {
     const container = document.getElementById('history-list');
     if (!records || records.length === 0) {
@@ -1104,3 +1136,319 @@ async function refreshSingleSymbol(symbol, btnElement = null) {
         btn.innerHTML = originalContent;
     }
 }
+
+// ==================== AI智能分析功能 ====================
+
+let currentAIAnalysis = null;
+
+async function runAIAnalysis(forceRefresh = false) {
+    if (!currentSymbol) {
+        showToast('warning', '提示', '请先选择一个品种');
+        return;
+    }
+    
+    const btn = document.getElementById('ai-analyze-btn');
+    const content = document.getElementById('ai-analysis-content');
+    
+    btn.disabled = true;
+    btn.innerHTML = '<i class="fas fa-spinner fa-spin"></i><span>分析中...</span>';
+    
+    content.innerHTML = `
+        <div class="ai-analysis-loading">
+            <i class="fas fa-brain"></i>
+            <span>AI正在分析中...</span>
+        </div>
+    `;
+    
+    try {
+        const response = await fetch(`${API_BASE}/ai-analysis/${currentSymbol}?force_refresh=${forceRefresh}`);
+        const data = await response.json();
+        
+        if (data.success) {
+            currentAIAnalysis = data.data;
+            displayAIAnalysisResult(data.data);
+            showToast('success', '分析完成', `${currentSymbol} AI分析已完成`);
+        } else {
+            content.innerHTML = `
+                <div class="ai-analysis-placeholder">
+                    <i class="fas fa-exclamation-triangle"></i>
+                    <p>${data.error || 'AI分析失败，请稍后重试'}</p>
+                </div>
+            `;
+            showToast('error', '分析失败', data.error || 'AI分析失败');
+        }
+    } catch (error) {
+        console.error('AI分析请求失败:', error);
+        content.innerHTML = `
+            <div class="ai-analysis-placeholder">
+                <i class="fas fa-exclamation-triangle"></i>
+                <p>网络错误，请检查网络连接</p>
+            </div>
+        `;
+        showToast('error', '请求失败', '网络错误，请稍后重试');
+    } finally {
+        btn.disabled = false;
+        btn.innerHTML = '<i class="fas fa-play"></i><span>智能分析</span>';
+    }
+}
+
+function displayAIAnalysisResult(data) {
+    const content = document.getElementById('ai-analysis-content');
+    const result = data.result;
+    const timestamp = new Date(data.analysis_time).toLocaleString('zh-CN');
+    
+    const direction = result.trading_suggestion?.direction || '观望';
+    const directionClass = direction === '做多' ? 'long' : direction === '做空' ? 'short' : 'neutral';
+    const directionIcon = direction === '做多' ? 'fa-arrow-up' : direction === '做空' ? 'fa-arrow-down' : 'fa-arrows-left-right';
+    const confidence = result.trading_suggestion?.confidence || 0;
+    
+    const entryMin = result.trading_suggestion?.entry_range?.min || '--';
+    const entryMax = result.trading_suggestion?.entry_range?.max || '--';
+    const stopLoss = result.trading_suggestion?.stop_loss || '--';
+    const positionSize = result.trading_suggestion?.position_size || '--';
+    
+    content.innerHTML = `
+        <div class="ai-analysis-result">
+            <div class="ai-summary">${result.summary || '暂无总结'}</div>
+            
+            <div class="ai-suggestion-row">
+                <div class="ai-suggestion-direction ${directionClass}">
+                    <i class="fas ${directionIcon}"></i>
+                    <span style="font-weight:600;">${direction}</span>
+                </div>
+                <div class="ai-confidence">
+                    <span>置信度</span>
+                    <div class="ai-confidence-bar">
+                        <div class="ai-confidence-fill" style="width: ${confidence}%"></div>
+                    </div>
+                    <span>${confidence}%</span>
+                </div>
+            </div>
+            
+            <div class="ai-key-metrics">
+                <div class="ai-metric-item">
+                    <span class="label">入场区间</span>
+                    <span class="value">${entryMin}-${entryMax}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">止损位</span>
+                    <span class="value down">${stopLoss}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">建议仓位</span>
+                    <span class="value">${positionSize}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">纪律评分</span>
+                    <span class="value">${result.discipline_score?.total || '--'}/${result.discipline_score?.max || '11'}</span>
+                </div>
+            </div>
+            
+            <div class="ai-timestamp">
+                <i class="fas fa-clock"></i> 分析时间: ${timestamp}
+                <button class="ai-detail-btn" onclick="showAIDetailModal()" style="margin-left: 8px; background: none; border: 1px solid var(--border-color); padding: 4px 12px; border-radius: 4px; color: var(--cyan); cursor: pointer; font-size: 11px;">
+                    查看详情 <i class="fas fa-external-link-alt"></i>
+                </button>
+            </div>
+        </div>
+    `;
+}
+
+function showAIDetailModal() {
+    if (!currentAIAnalysis) {
+        showToast('warning', '提示', '暂无AI分析数据');
+        return;
+    }
+    
+    const result = currentAIAnalysis.result;
+    const modalBody = document.getElementById('ai-analysis-modal-body');
+    
+    let fourDimensionalHTML = '';
+    const periods = ['60min', '30min', '15min'];
+    periods.forEach(period => {
+        const data = result.four_dimensional?.[period];
+        if (data) {
+            fourDimensionalHTML += `
+                <tr>
+                    <td class="period-cell">${period}</td>
+                    <td>
+                        <div>趋势: ${data.macd?.trend || '--'}</div>
+                        <div>位置: ${data.macd?.position || '--'}</div>
+                        <div>柱状图: ${data.macd?.histogram || '--'}</div>
+                    </td>
+                    <td>
+                        <div>状态: ${data.volume?.status || '--'}</div>
+                        <div>量比: ${data.volume?.ratio || '--'}</div>
+                    </td>
+                    <td>
+                        <div>K: ${data.kdj?.k || '--'} D: ${data.kdj?.d || '--'}</div>
+                        <div>信号: ${data.kdj?.signal || '--'}</div>
+                        <div>状态: ${data.kdj?.status || '--'}</div>
+                    </td>
+                    <td>${data.conclusion || '--'}</td>
+                </tr>
+            `;
+        }
+    });
+    
+    let scenariosHTML = '';
+    const scenarios = result.scenario_plans || {};
+    const scenarioIcons = {
+        'breakthrough': 'fa-rocket',
+        'consolidation': 'fa-exchange-alt',
+        'reversal': 'fa-undo',
+        'news_impact': 'fa-newspaper'
+    };
+    
+    Object.entries(scenarios).forEach(([key, scenario]) => {
+        scenariosHTML += `
+            <div class="scenario-card ${key}">
+                <div class="scenario-probability">${scenario.probability || 0}%</div>
+                <div class="scenario-action">${scenario.action || '--'}</div>
+            </div>
+        `;
+    });
+    
+    let redLinesHTML = '';
+    if (result.red_lines_check?.violated?.length > 0) {
+        result.red_lines_check.violated.forEach(line => {
+            redLinesHTML += `<div class="red-line-item"><i class="fas fa-exclamation-circle"></i> ${line}</div>`;
+        });
+    } else {
+        redLinesHTML = '<div class="red-line-item pass"><i class="fas fa-check-circle"></i> 未触碰交易红线</div>';
+    }
+    
+    let disciplineHTML = '';
+    const discipline = result.discipline_score?.details || {};
+    const disciplineLabels = {
+        'trend': '趋势',
+        'position': '位置',
+        'signal': '信号',
+        'risk': '风险',
+        'mindset': '心态'
+    };
+    
+    Object.entries(disciplineLabels).forEach(([key, label]) => {
+        const isPass = discipline[key];
+        disciplineHTML += `
+            <div class="discipline-item ${isPass ? 'pass' : 'fail'}">
+                <i class="fas ${isPass ? 'fa-check-circle' : 'fa-times-circle'}"></i>
+                <span class="discipline-label">${label}</span>
+            </div>
+        `;
+    });
+    
+    let experiencesHTML = '';
+    const experiences = result.experience_lessons || [];
+    if (experiences.length > 0) {
+        experiences.forEach(exp => {
+            experiencesHTML += `<div class="experience-item"><i class="fas fa-lightbulb"></i> ${exp}</div>`;
+        });
+    } else {
+        experiencesHTML = '<div class="experience-item"><i class="fas fa-info-circle"></i> 暂无经验提醒</div>';
+    }
+    
+    modalBody.innerHTML = `
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-chart-bar"></i>
+                四维联合信号表 (4D-XV)
+            </div>
+            <table class="four-dimensional-table">
+                <thead>
+                    <tr>
+                        <th>周期</th>
+                        <th>MACD(趋势)</th>
+                        <th>成交量(资金)</th>
+                        <th>KDJ(时机)</th>
+                        <th>结论</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    ${fourDimensionalHTML || '<tr><td colspan="5">暂无数据</td></tr>'}
+                </tbody>
+            </table>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-crosshairs"></i>
+                关键点位 (Pivot Point)
+            </div>
+            <div class="ai-key-metrics">
+                <div class="ai-metric-item">
+                    <span class="label">R2</span>
+                    <span class="value down">${result.pivot_points?.r2 || '--'}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">R1</span>
+                    <span class="value down">${result.pivot_points?.r1 || '--'}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">PP</span>
+                    <span class="value" style="color: var(--purple);">${result.pivot_points?.pp || '--'}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">S1</span>
+                    <span class="value up">${result.pivot_points?.s1 || '--'}</span>
+                </div>
+                <div class="ai-metric-item">
+                    <span class="label">S2</span>
+                    <span class="value up">${result.pivot_points?.s2 || '--'}</span>
+                </div>
+            </div>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-balance-scale"></i>
+                交易红线审查
+            </div>
+            <div class="red-lines-list">
+                ${redLinesHTML}
+            </div>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-check-double"></i>
+                11项纪律检查
+            </div>
+            <div class="discipline-grid">
+                ${disciplineHTML}
+            </div>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-project-diagram"></i>
+                情景预案
+            </div>
+            <div class="scenario-cards">
+                ${scenariosHTML || '<div class="scenario-card"><div class="scenario-action">暂无情景预案</div></div>'}
+            </div>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-lightbulb"></i>
+                经验教训提醒
+            </div>
+            <div class="ai-experience-list">
+                ${experiencesHTML}
+            </div>
+        </div>
+        
+        <div class="ai-modal-section">
+            <div class="ai-modal-section-title">
+                <i class="fas fa-exclamation-triangle"></i>
+                风险提示
+            </div>
+            <div class="ai-experience-list">
+                ${(result.risk_warnings || []).map(w => `<div class="experience-item"><i class="fas fa-exclamation-circle"></i> ${w}</div>`).join('')}
+            </div>
+        </div>
+    `;
+    
+    document.getElementById('ai-analysis-modal').classList.add('active');
+}

check_db.py

@@ -0,0 +1,55 @@
+import sqlite3
+import os
+from pathlib import Path
+
+db_path = Path(__file__).parent / "data" / "futures_analysis.db"
+
+print("=" * 60)
+print("数据库信息检查")
+print("=" * 60)
+print(f"数据库路径: {db_path}")
+print(f"文件存在: {db_path.exists()}")
+
+if db_path.exists():
+    print(f"文件大小: {db_path.stat().st_size} bytes")
+    print(f"最后修改: {db_path.stat().st_mtime}")
+
+    conn = sqlite3.connect(str(db_path))
+    cursor = conn.cursor()
+
+    # 获取所有表
+    cursor.execute("SELECT name FROM sqlite_master WHERE type='table'")
+    tables = cursor.fetchall()
+    print(f"\n数据库表 ({len(tables)}):")
+    for table in tables:
+        table_name = table[0]
+        cursor.execute(f"SELECT COUNT(*) FROM {table_name}")
+        count = cursor.fetchone()[0]
+        print(f"  - {table_name}: {count} 条记录")
+
+    # 检查AI模型配置
+    print("\n" + "=" * 60)
+    print("AI模型配置检查")
+    print("=" * 60)
+    try:
+        cursor.execute("SELECT id, provider, model_name, is_active, enabled FROM ai_model_configs")
+        models = cursor.fetchall()
+        print(f"AI模型数量: {len(models)}")
+        if models:
+            for m in models:
+                print(f"  ID: {m[0]}")
+                print(f"  Provider: {m[1]}")
+                print(f"  Model Name: {m[2]}")
+                print(f"  Active: {m[3]}")
+                print(f"  Enabled: {m[4]}")
+                print()
+        else:
+            print("  没有配置AI模型!")
+    except sqlite3.OperationalError as e:
+        print(f"  查询失败: {e}")
+
+    conn.close()
+else:
+    print("数据库文件不存在!")
+
+print("=" * 60)

data/ai_analysis_prompt.md

@@ -0,0 +1,201 @@
+# 📋 AI 期货/股票四维联合分析系统 (Prompt 模板)
+
+> **版本**: v2.0 (全量实战版)  
+> **核心架构**: 四维联合判断 (4D-XV) + 17 条交易红线 + 多周期共振 + 动态策略切换  
+> **适用场景**: 期货/股票技术分析、日内/短线交易决策、交易纪律审查
+
+---
+
+## 🧠 角色与任务设定 (System Prompt)
+
+```text
+你是一位拥有 20 年实战经验的资深金融交易分析师，精通 A 股市场与商品期货的技术分析。
+你的核心使命是基于提供的 K 线数据，执行**「四维联合判断分析法 (4D-XV)」**，并提供包含风控红线审查的客观交易策略。
+
+你的分析必须遵循以下原则：
+1. **数据驱动**：所有结论必须基于输入的 JSON 数据，严禁凭空捏造。
+2. **四维共振**：任何交易建议必须经过 MACD（趋势）、成交量（资金）、KDJ（时机）、多周期（方向）的交叉验证。
+3. **红线否决**：如果数据触发「17 条交易红线」，必须直接给出「禁止交易」或「止损」的建议。
+4. **客观中立**：不使用绝对化表述，提供情景预案（概率估算）。
+5. **动态切换**：遇到关键位放量突破时，立即切换右侧思维，不逆势死扛。
+```
+
+---
+
+## 📥 输入数据规范 (JSON)
+
+AI 模型接收的输入数据必须包含以下结构：
+
+```json
+{
+  "symbol": "AG2606",
+  "current_price": 7200,
+  "timeframes": {
+    "60min": {
+      "trend": "down",
+      "ma10": 7250, "ma20": 7300,
+      "macd_dif": -5.2, "macd_dea": -3.1, "macd_histogram": -2.1,
+      "kdj_k": 85, "kdj_d": 80, "kdj_j": 95,
+      "volume_avg": 150000, "volume_current": 200000,
+      "candles": [{"time": "...", "open": 0, "high": 0, "low": 0, "close": 0, "volume": 0}]
+    },
+    "30min": { "... 结构同上 ..." },
+    "15min": { "... 结构同上 ..." },
+    "5min": { "... 结构同上 ..." },
+    "daily": { "... 结构同上 ..." }
+  },
+  "fundamentals": "近期库存持续下降，宏观情绪偏暖"
+}
+```
+
+---
+
+## 🛠️ 核心分析引擎 (执行步骤)
+
+请在代码中要求 AI 严格按以下步骤执行分析：
+
+### 第一步：四维联合信号扫描 (4D-XV)
+对 **60m/30m/15m** 周期进行四维交叉验证，填写诊断表：
+
+| 周期 | MACD(趋势锚点) | 成交量(资金验证) | KDJ(时机狙击) | 四维综合结论 |
+|------|----------------|------------------|---------------|--------------|
+| **60m** | 零轴上/下，柱状图放大/缩小/背离 | 趋势量能（放量上涨/缩量回调等） | K 值位置，金死叉状态，是否钝化 | 定大势（只做多/只做空/震荡） |
+| **30m** | 零轴上/下，柱状图放大/缩小 | 拐点量能（突破/跌破量比≥1.5） | K 值位置，金死叉状态 | 找拐点（验证短期转折） |
+| **15m** | 零轴上/下，柱状图放大/缩小 | 入场量能（即时量比≥1.5/2.0） | K 值位置，金死叉状态 | 择入场（精确触发点） |
+
+**判定标准**：
+*   ✅ **完美共振**：3 维度以上同向 + 量能确认 → 强信号（胜率≥80%）
+*   ⚠️ **弱势/矛盾**：指标矛盾或缺少量能验证 → 震荡/观望/轻仓
+*   ❌ **逆势/背离**：量价背离、MACD 与 KDJ 冲突 → **禁止入场**
+
+### 第二步：KDJ 深度诊断
+*   **区域判断**：超买 (>80)、超卖 (<20)、中性区。
+*   **信号确认**：金叉/死叉有效性（需等待第二次确认，防假信号）。
+*   **钝化检查**：是否持续超买/超卖 >6 根 K 线？若是，**放弃 KDJ 逆势信号，改用 MACD 趋势跟踪**。
+*   **背离识别**：KDJ 顶/底背离 + MACD 同步背离 = 极高胜率反转信号。
+*   **参数建议**：日线 (9,3,3)、60m (9,3,3)、30m (6,3,3)、15m (5,3,3)。
+
+### 第三步：关键位计算与验证 (Pivot Point)
+使用最近一根完整 K 线的 H、L、C 计算：
+`PP = (H+L+C)/3`, `R1 = 2*PP-L`, `S1 = 2*PP-H`, `R2 = PP+(H-L)`, `S2 = PP-(H-L)`
+
+**关键位五维度验证**：
+1.  测试次数 ≥3 次？
+2.  量价配合良好（无量支撑为虚位）？
+3.  多周期共振（大周期关键位权重更高）？
+4.  基本面匹配（成本线、政策线）？
+5.  突破后回踩确认？
+
+### 第四步：交易红线审查 (17 条)
+**检查数据是否触碰以下红线（触碰即停）**：
+*   [ ] 逆势重仓
+*   [ ] 移动止损/加仓摊平
+*   [ ] 情绪化交易/报复交易
+*   [ ] 放量突破死扛逆势单（最高红线）
+*   [ ] 22:00 后开仓/14:30 后隔夜决策
+*   [ ] 无衰竭信号入场/追涨杀跌
+*   [ ] 五维度评分 < 9 分 或 衰竭评分 < 5 分
+
+### 第五步：动态策略切换 (Plan B)
+*   **场景**：价格逼近关键阻力位，原判断受阻回落，但出现**放量突破**。
+*   **验证**：15m/30m 成交量 > 近 5 日均量 200%，且伴随实体饱满大阳线。
+*   **动作**：若持有逆势单（如空单），价格站稳 15 分钟 → **无条件市价止损**，绝不等回调。
+*   **思维**：立即切换为右侧跟随思维，回踩阻力位（转支撑）时尝试做多。
+
+---
+
+## 📝 输出模板 (Markdown)
+
+要求 AI 严格按照以下格式输出：
+
+```markdown
+## 📊 {symbol} 四维联合分析报告
+**分析时间**：{当前时间} | **当前价**：{current_price}
+
+### 一、四维联合信号表 (4D-XV)
+| 周期 | MACD(趋势) | 成交量(资金) | KDJ(时机) | 结论 |
+|------|------------|-------------|-----------|------|
+| 60m | ... | ... | ... | ... |
+| 30m | ... | ... | ... | ... |
+| 15m | ... | ... | ... | ... |
+
+### 二、KDJ 深度诊断
+- **当前状态**：...
+- **钝化/背离**：...
+- **参数建议**：...
+
+### 三、关键位 (Pivot Point)
+- 🔴 **R2**: ... | **R1**: ...
+- ⚪ **PP**: ...
+- 🟢 **S1**: ... | **S2**: ...
+- **有效性验证**：(测试次数、量能、周期共振情况)
+
+### 四、11 项纪律检查与红线审查
+- **红线检查**：✅ 未触碰 / ❌ 触碰 [具体红线]
+- **纪律得分**：__ / 11 (≥9 可入场)
+- **诊断表**：
+    - A. 趋势 (60m 大势/30m 结构/15m 协同): [✅/❌]
+    - B. 位置 (关键位/共振): [✅/❌]
+    - C. 信号 (K 线/量能/背离): [✅/❌]
+    - D. 风险 (盈亏比/仓位): [✅/❌]
+    - E. 心态: [✅/❌]
+
+### 五、交易建议
+- **方向**：做多 / 做空 / 观望
+- **入场区间**：...
+- **止损位**：... (依据：前低/高/关键位)
+- **止盈目标**：... (盈亏比 ≥ 1:2)
+- **建议仓位**：... (基于信号强度)
+
+### 六、动态策略预案 (Plan B)
+1. **顺势突破 (35-45%)**：放量突破关键位，跟随右侧思维，移动止损。
+2. **震荡拉锯 (30-40%)**：R1-S1 区间内高抛低吸，突破前观望。
+3. **假突破反转 (15-25%)**：MACD 顶/底背离 + 量能不足，立即止损反手。
+4. **突发消息 (5-10%)**：减仓 50% 规避不确定性。
+
+### ⚠️ 风险提示与经验教训
+- **技术局限**：技术指标具有滞后性，历史表现不代表未来。
+- **避坑提醒**：(根据当前盘面自动提示，如"警惕缩量创新高"、"KDJ 超买钝化中不宜逆势做空"等)
+- **红线警告**：(如触发红线，在此处高亮警告)
+- **免责声明**：本分析仅供参考，不构成投资建议，交易需自负盈亏。
+```
+
+---
+
+## 💡 经验教训总结 (供 AI 内部参考)
+
+在分析过程中，AI 应时刻检索以下**历史经验教训库**，若数据特征匹配，必须在报告中提示：
+
+| 陷阱名称 | 特征识别 | 应对经验教训 |
+|----------|----------|--------------|
+| **唯 MACD 论** | MACD 金叉但 KDJ 超买 | 往往是诱多，**禁止单独使用 MACD 入场**。 |
+| **唯 KDJ 论** | KDJ 金叉但 MACD 零轴下死叉 | 通常是下跌中继（假反弹），**必须等待 MACD 确认**。 |
+| **无量突破** | KDJ+MACD 共振但无量 | 假突破概率 > 80%，**放弃或极轻仓试错**。 |
+| **大周期否决** | 日线/60m 明确空头 | 小周期任何做多信号均为短线反弹，**仓位必须减半**。 |
+| **钝化区逆势** | KDJ 超买但价格继续涨 | 不逆势！等钝化解除（KDJ 离开超买区 + 价格反向突破）。 |
+| **假金叉/死叉** | 金叉后快速死叉 | 等待**第二次确认**，或结合 30m 级别过滤。 |
+| **背离假信号** | 多次背离后价格继续原方向 | 背离不是入场理由，**等待价格突破关键位确认**。 |
+| **放量突破死扛** | 价格放量突破阻力，仍持有逆势空单 | **最高红线！立即市价止损，绝不等回调。** |
+
+---
+
+## 📦 附录：精简版 Prompt (上下文窗口有限时使用)
+
+```text
+你是一位专业交易分析师。请基于以下 JSON 数据，执行【四维联合判断 (4D-XV)】分析。
+
+分析逻辑：
+1. MACD 定趋势，KDJ 择时机，成交量验证，多周期 (60m→30m→15m) 共振。
+2. 严格审查「17 条交易红线」，触碰即禁止交易。
+3. 若出现放量突破关键位，执行 Plan B（右侧跟随，不逆势死扛）。
+
+输出内容：
+1. 四维信号表 (MACD/量/KDJ/结论)。
+2. Pivot Point 关键位及有效性验证。
+3. 11 项纪律检查评分。
+4. 交易建议 (方向/入场/止损/止盈/仓位)。
+5. 情景预案 (5 种情景 + 概率)。
+6. 风险提示与避坑提醒 (结合经验教训库)。
+
+数据：{{JSON_DATA}}
+```

init_analysis_db.py

@@ -0,0 +1,24 @@
+"""
+手动初始化期货智析数据库
+用于在应用外部创建数据库表
+"""
+import sys
+from pathlib import Path
+
+# 添加项目根目录到 Python 路径
+project_root = Path(__file__).parent
+sys.path.insert(0, str(project_root))
+
+from app.analysis_db import init_analysis_db
+
+if __name__ == "__main__":
+    print("开始初始化期货智析数据库...")
+    init_analysis_db()
+    print("✅ 数据库初始化完成！")
+    
+    # 验证数据库文件是否创建
+    from app.analysis_db import ANALYSIS_DB_PATH
+    if ANALYSIS_DB_PATH.exists():
+        print(f"✅ 数据库文件已创建: {ANALYSIS_DB_PATH}")
+    else:
+        print(f"❌ 数据库文件未创建: {ANALYSIS_DB_PATH}")

verify_db.py

@@ -0,0 +1,27 @@
+"""
+验证数据库表是否正确创建
+"""
+import sqlite3
+
+db_path = "data/futures_analysis.db"
+
+conn = sqlite3.connect(db_path)
+cursor = conn.execute('SELECT name FROM sqlite_master WHERE type="table"')
+
+print("✅ 数据库表列表:")
+for row in cursor:
+    print(f"  - {row[0]}")
+
+# 检查 ai_analysis_cache 表是否存在
+cursor.execute('SELECT name FROM sqlite_master WHERE type="table" AND name="ai_analysis_cache"')
+if cursor.fetchone():
+    print("\n✅ ai_analysis_cache 表已存在！")
+    # 显示表结构
+    cursor.execute("PRAGMA table_info(ai_analysis_cache)")
+    print("\n表结构:")
+    for col in cursor:
+        print(f"  {col[1]} ({col[2]})")
+else:
+    print("\n❌ ai_analysis_cache 表不存在！")
+
+conn.close()