1

backend/app/analysis/breakout_signals.py

@@ -0,0 +1,648 @@
+"""趋势突破入场信号分类 + 供需分析 + 趋势评分
+
+三种入场信号类型：
+- 突破型 (breakout): 放量突破 20 日阻力位
+- 回踩型 (pullback): 上升趋势中缩量回踩均线支撑
+- 启动型 (launch): 高位缩量整理后即将变盘
+"""
+
+import logging
+from enum import Enum
+
+import numpy as np
+import pandas as pd
+
+from app.config import settings
+
+logger = logging.getLogger(__name__)
+
+
+class EntrySignal(str, Enum):
+    BREAKOUT = "breakout"
+    PULLBACK = "pullback"
+    LAUNCH = "launch"
+    NONE = "none"
+
+
+# ── 入场信号检测 ──
+
+
+def classify_entry_signal(df: pd.DataFrame) -> dict:
+    """分类入场信号类型
+
+    优先级: breakout > pullback > launch
+
+    Returns:
+        {
+            "signal_type": EntrySignal,
+            "signal_score": float (0-100),
+            "details": dict,  # 信号相关详情
+        }
+    """
+    if df is None or len(df) < 30:
+        return {"signal_type": EntrySignal.NONE, "signal_score": 0, "details": {}}
+
+    # 优先检测突破型（最符合 1-5 天操作窗口）
+    result = detect_breakout(df)
+    if result:
+        result["signal_score"] = _score_breakout(df, result)
+        return result
+
+    # 回踩型（风险收益比最好）
+    result = detect_pullback(df)
+    if result:
+        result["signal_score"] = _score_pullback(df, result)
+        return result
+
+    # 启动型（需要耐心等待）
+    result = detect_launch(df)
+    if result:
+        result["signal_score"] = _score_launch(df, result)
+        return result
+
+    return {"signal_type": EntrySignal.NONE, "signal_score": 0, "details": {}}
+
+
+def detect_breakout(df: pd.DataFrame) -> dict | None:
+    """突破型检测
+
+    条件（全部满足）：
+    1. 价格突破 20 日阻力位
+    2. 量能放大 (vol > vol_ma5 * 1.5)
+    3. 短期均线多头 (MA5 > MA10 > MA20)
+    4. 收盘偏强 (close 在当日区间上半部分)
+    """
+    last = df.iloc[-1]
+
+    # 必需指标列
+    required = ["close", "high", "low", "vol", "ma5", "ma10", "ma20", "vol_ma5"]
+    if any(c not in df.columns for c in required):
+        return None
+    if pd.isna(last["vol_ma5"]) or last["vol_ma5"] == 0:
+        return None
+
+    # 1. 突破 20 日阻力（允许距阻力位 1% 以内，贴近即视为突破前兆）
+    if len(df) < 22:
+        return None
+    resist_20d = df["high"].iloc[-21:-1].max()
+    if last["close"] < resist_20d * 0.99:
+        return None
+
+    # 2. 量能放大
+    min_vol_ratio = getattr(settings, "breakout_min_volume_ratio", 1.2)
+    if last["vol"] <= last["vol_ma5"] * min_vol_ratio:
+        return None
+
+    # 3. 短期均线偏多（MA5 > MA20 即可，不要求三线完美排列）
+    if not (last["ma5"] > last["ma20"]):
+        return None
+
+    # 4. 收盘偏强（不低于当日中点）
+    day_range = last["high"] - last["low"]
+    if day_range > 0 and (last["close"] - last["low"]) / day_range < 0.5:
+        return None
+
+    breakout_pct = (last["close"] - resist_20d) / resist_20d * 100
+
+    return {
+        "signal_type": EntrySignal.BREAKOUT,
+        "details": {
+            "resist_level": round(resist_20d, 2),
+            "breakout_pct": round(breakout_pct, 2),
+            "volume_ratio": round(last["vol"] / last["vol_ma5"], 2),
+        },
+    }
+
+
+def detect_pullback(df: pd.DataFrame) -> dict | None:
+    """回踩型检测
+
+    条件（全部满足）：
+    1. MA20 > MA60 — 中长期上升趋势
+    2. 近 5 日内有 MA5 > MA20 — 近期处于上升趋势
+    3. 价格靠近 MA10 或 MA20 (<2%)
+    4. 缩量回调 (近 3 日均量 < 前 5 日均量 * 0.7)
+    5. RSI 40-65
+    """
+    last = df.iloc[-1]
+
+    required = ["close", "vol", "ma5", "ma10", "ma20", "ma60", "rsi14"]
+    if any(c not in df.columns for c in required):
+        return None
+    if pd.isna(last["ma60"]) or pd.isna(last["rsi14"]):
+        return None
+
+    # 1. 中长期上升趋势
+    if not (last["ma20"] > last["ma60"]):
+        return None
+
+    # 2. 近 5 日内有 MA5 > MA20
+    recent_5 = df.tail(5)
+    has_uptrend = any(
+        row["ma5"] > row["ma20"]
+        for _, row in recent_5.iterrows()
+        if not pd.isna(row["ma5"]) and not pd.isna(row["ma20"])
+    )
+    if not has_uptrend:
+        return None
+
+    # 3. 价格靠近 MA10 或 MA20（放宽至 4%）
+    near_ma10 = (
+        not pd.isna(last["ma10"])
+        and last["ma10"] > 0
+        and abs(last["close"] - last["ma10"]) / last["ma10"] < 0.04
+    )
+    near_ma20 = (
+        not pd.isna(last["ma20"])
+        and last["ma20"] > 0
+        and abs(last["close"] - last["ma20"]) / last["ma20"] < 0.04
+    )
+    if not (near_ma10 or near_ma20):
+        return None
+
+    # 4. 缩量回调（放宽至 0.85）
+    if len(df) < 8:
+        return None
+    recent_3_vol = df["vol"].tail(3).mean()
+    prev_5_vol = df["vol"].iloc[-8:-3].mean()
+    if prev_5_vol == 0:
+        return None
+    shrink_ratio = recent_3_vol / prev_5_vol
+    max_shrink = getattr(settings, "pullback_max_shrink_ratio", 0.85)
+    if shrink_ratio >= max_shrink:
+        return None
+
+    # 5. RSI 范围（放宽至 35-70）
+    if not (35 <= last["rsi14"] <= 70):
+        return None
+
+    support_ma = "MA10" if near_ma10 else "MA20"
+    support_price = last["ma10"] if near_ma10 else last["ma20"]
+
+    return {
+        "signal_type": EntrySignal.PULLBACK,
+        "details": {
+            "support_ma": support_ma,
+            "support_price": round(support_price, 2),
+            "volume_shrink_ratio": round(shrink_ratio, 2),
+        },
+    }
+
+
+def detect_launch(df: pd.DataFrame) -> dict | None:
+    """启动型检测（高位缩量整理后即将变盘）
+
+    条件（全部满足）：
+    1. close > MA60 且距 20 日高点 < 5%
+    2. 近 10 日振幅 < 5%，每日涨跌幅 < 3%
+    3. 近 5 日均量 < 前 5 日均量 * 0.6
+    4. MA5 > MA20（多头格局未破）
+    5. 布林带宽收窄
+    6. DIF > 0
+    """
+    last = df.iloc[-1]
+
+    required = ["close", "high", "vol", "pct_chg", "ma5", "ma20", "ma60",
+                "dif", "boll_bw"]
+    if any(c not in df.columns for c in required):
+        return None
+    if pd.isna(last["ma60"]) or pd.isna(last["dif"]) or pd.isna(last["boll_bw"]):
+        return None
+
+    # 1. 高位
+    if last["close"] < last["ma60"]:
+        return None
+    high_20d = df["high"].tail(20).max()
+    if (high_20d - last["close"]) / high_20d > 0.05:
+        return None
+
+    # 2. 窄幅整理（放宽振幅至 8%，每日涨跌幅至 4%）
+    if len(df) < 10:
+        return None
+    recent = df.tail(10)
+    close_range = (recent["close"].max() - recent["close"].min()) / recent["close"].min()
+    max_range = getattr(settings, "consolidation_max_range_pct", 8.0) / 100
+    if close_range > max_range:
+        return None
+    if any(abs(row["pct_chg"]) > 4 for _, row in recent.iterrows() if not pd.isna(row.get("pct_chg"))):
+        return None
+
+    # 3. 缩量（放宽至 0.75）
+    if len(df) < 15:
+        return None
+    recent_5_vol = df["vol"].tail(5).mean()
+    earlier_5_vol = df["vol"].iloc[-15:-10].mean()
+    if earlier_5_vol == 0:
+        return None
+    vol_ratio = recent_5_vol / earlier_5_vol
+    if vol_ratio >= 0.75:
+        return None
+
+    # 4. 多头格局未破
+    if last["ma5"] <= last["ma20"]:
+        return None
+
+    # 5-6. 布林收窄 + MACD 零轴上方（软条件，不满足不淘汰）
+    boll_narrowing = False
+    dif_positive = False
+
+    if len(df) >= 20 and not pd.isna(last["boll_bw"]):
+        bw_10d_ago = df.iloc[-10]["boll_bw"]
+        if not pd.isna(bw_10d_ago) and last["boll_bw"] < bw_10d_ago:
+            boll_narrowing = True
+
+    if not pd.isna(last["dif"]) and last["dif"] > 0:
+        dif_positive = True
+
+    return {
+        "signal_type": EntrySignal.LAUNCH,
+        "details": {
+            "consolidation_days": 10,
+            "volume_shrink_ratio": round(vol_ratio, 2),
+            "price_range_pct": round(close_range * 100, 2),
+            "resist_level": round(high_20d, 2),
+            "boll_narrowing": boll_narrowing,
+            "dif_positive": dif_positive,
+        },
+    }
+
+
+# ── 信号质量评分 ──
+
+
+def _score_breakout(df: pd.DataFrame, signal: dict) -> float:
+    """突破型信号质量评分 (0-100)"""
+    last = df.iloc[-1]
+    score = 0
+
+    # 突破力度 (25)
+    breakout_pct = signal["details"].get("breakout_pct", 0)
+    if breakout_pct > 3:
+        score += 25
+    elif breakout_pct > 1.5:
+        score += 18
+    elif breakout_pct > 0.5:
+        score += 10
+
+    # 量能放大程度 (25)
+    vol_ratio = signal["details"].get("volume_ratio", 1)
+    if vol_ratio > 2.5:
+        score += 25
+    elif vol_ratio > 2.0:
+        score += 20
+    elif vol_ratio > 1.5:
+        score += 12
+
+    # 均线完整性 (25)
+    if all(c in df.columns for c in ["ma60"]):
+        if not pd.isna(last["ma60"]) and last["ma5"] > last["ma10"] > last["ma20"] > last["ma60"]:
+            score += 25
+        elif last["ma5"] > last["ma10"] > last["ma20"]:
+            score += 15
+    else:
+        score += 10
+
+    # MACD 状态 (15)
+    if all(c in df.columns for c in ["dif", "dea"]):
+        if not pd.isna(last["dif"]) and last["dif"] > last["dea"] and last["dif"] > 0:
+            score += 15
+        elif not pd.isna(last["dif"]) and last["dif"] > last["dea"]:
+            score += 8
+
+    # RSI 区间 (10)
+    if "rsi14" in df.columns and not pd.isna(last["rsi14"]):
+        if 50 <= last["rsi14"] <= 75:
+            score += 10
+        elif 45 <= last["rsi14"] <= 80:
+            score += 5
+
+    return min(score, 100)
+
+
+def _score_pullback(df: pd.DataFrame, signal: dict) -> float:
+    """回踩型信号质量评分 (0-100)"""
+    last = df.iloc[-1]
+    score = 0
+
+    # 缩量程度 (30)
+    shrink = signal["details"].get("volume_shrink_ratio", 1)
+    if shrink < 0.4:
+        score += 30
+    elif shrink < 0.5:
+        score += 22
+    elif shrink < 0.6:
+        score += 15
+
+    # 支撑有效性 (25)
+    support_ma = signal["details"].get("support_ma", "")
+    if support_ma == "MA20":
+        score += 25  # MA20 支撑更强
+    elif support_ma == "MA10":
+        score += 15
+
+    # 均线多头 (25)
+    if all(c in df.columns for c in ["ma60"]):
+        if not pd.isna(last["ma60"]) and last["ma5"] > last["ma10"] > last["ma20"] > last["ma60"]:
+            score += 25
+        elif last["ma5"] > last["ma10"] > last["ma20"]:
+            score += 15
+
+    # MACD 状态 (10)
+    if all(c in df.columns for c in ["dif", "dea"]):
+        if not pd.isna(last["dif"]) and last["dif"] > 0:
+            score += 10
+        elif not pd.isna(last["dif"]) and last["dif"] > last["dea"]:
+            score += 5
+
+    # RSI (10)
+    if "rsi14" in df.columns and not pd.isna(last["rsi14"]):
+        if 45 <= last["rsi14"] <= 60:
+            score += 10
+        elif 40 <= last["rsi14"] <= 65:
+            score += 5
+
+    return min(score, 100)
+
+
+def _score_launch(df: pd.DataFrame, signal: dict) -> float:
+    """启动型信号质量评分 (0-100)"""
+    last = df.iloc[-1]
+    score = 0
+    details = signal.get("details", {})
+
+    # 整理充分度 (25) — 振幅越小越好
+    range_pct = details.get("price_range_pct", 10)
+    if range_pct < 2:
+        score += 25
+    elif range_pct < 4:
+        score += 18
+    elif range_pct < 8:
+        score += 10
+
+    # 缩量程度 (25)
+    shrink = details.get("volume_shrink_ratio", 1)
+    if shrink < 0.3:
+        score += 25
+    elif shrink < 0.5:
+        score += 18
+    elif shrink < 0.75:
+        score += 10
+
+    # 均线保持 (20)
+    if all(c in df.columns for c in ["ma60"]):
+        if not pd.isna(last["ma60"]) and last["ma5"] > last["ma20"] > last["ma60"]:
+            score += 20
+        elif last["ma5"] > last["ma20"]:
+            score += 10
+
+    # 布林收窄 (15) — 软条件，满足加分
+    if details.get("boll_narrowing"):
+        if "boll_bw" in df.columns and len(df) >= 20:
+            bw_now = last["boll_bw"]
+            bw_10d = df.iloc[-10]["boll_bw"]
+            if not pd.isna(bw_10d) and bw_now > 0:
+                narrowing = (bw_10d - bw_now) / bw_10d * 100
+                if narrowing > 30:
+                    score += 15
+                elif narrowing > 15:
+                    score += 10
+                else:
+                    score += 5
+
+    # MACD (15) — 软条件，满足加分
+    if details.get("dif_positive"):
+        score += 15
+    elif all(c in df.columns for c in ["dif"]) and not pd.isna(last.get("dif")):
+        score += 5
+
+    return min(score, 100)
+
+
+# ── 趋势 & 时机评分 ──
+
+
+def score_trend_timing(df: pd.DataFrame, entry_signal: dict) -> float:
+    """趋势&时机评分 (0-100)
+
+    维度：
+    - MA 排列 (25)
+    - MA20 方向 (10)
+    - 更高高点形态 (15)
+    - 入场信号质量 (50)
+    """
+    last = df.iloc[-1]
+    score = 0
+
+    # MA 排列 (25)
+    ma_cols = [c for c in ["ma5", "ma10", "ma20", "ma60"] if c in df.columns]
+    if len(ma_cols) >= 4 and not any(pd.isna(last[c]) for c in ma_cols):
+        if last["ma5"] > last["ma10"] > last["ma20"] > last["ma60"]:
+            score += 25
+        elif last["ma5"] > last["ma10"] > last["ma20"]:
+            score += 18
+        elif last["ma5"] > last["ma20"]:
+            score += 10
+
+    # MA20 方向 (10)
+    if "ma20" in df.columns and len(df) >= 3:
+        if not pd.isna(last["ma20"]) and not pd.isna(df.iloc[-3]["ma20"]):
+            if last["ma20"] > df.iloc[-3]["ma20"]:
+                score += 10
+
+    # 更高高点形态 (15)
+    if _check_higher_highs(df):
+        score += 15
+
+    # 入场信号质量 (50)
+    signal_score = entry_signal.get("signal_score", 0)
+    score += signal_score * 0.5
+
+    return min(score, 100)
+
+
+# ── 供需分析评分 ──
+
+
+def score_supply_demand(df: pd.DataFrame) -> float:
+    """供需评分 (0-100)
+
+    维度：
+    - 上涨日 vs 下跌日量比 (50): 需求主导度
+    - 量能收缩/扩张 (30): 供需变化趋势
+    - 量价配合 (20): 上涨放量+下跌缩量
+    """
+    score = 0
+
+    if len(df) < 10:
+        return 50.0
+
+    recent = df.tail(10)
+    last = df.iloc[-1]
+
+    # 上涨日 vs 下跌日量比 (50)
+    up_days = recent[recent["pct_chg"] > 0]
+    down_days = recent[recent["pct_chg"] <= 0]
+
+    if len(up_days) > 0 and len(down_days) > 0:
+        avg_up_vol = up_days["vol"].mean()
+        avg_down_vol = down_days["vol"].mean()
+        if avg_down_vol > 0:
+            ratio = avg_up_vol / avg_down_vol
+            if ratio > 2.0:
+                score += 50
+            elif ratio > 1.5:
+                score += 38
+            elif ratio > 1.2:
+                score += 25
+            elif ratio > 1.0:
+                score += 12
+    elif len(up_days) > 0:
+        score += 40  # 全部上涨日，需求强劲
+
+    # 量能收缩/扩张 (30)
+    vol_ma_col = "vol_ma10" if "vol_ma10" in df.columns else "vol_ma5"
+    if vol_ma_col in df.columns and not pd.isna(last[vol_ma_col]) and last[vol_ma_col] > 0:
+        recent_3_vol = recent["vol"].tail(3).mean()
+        shrink_ratio = recent_3_vol / last[vol_ma_col]
+
+        if shrink_ratio < 0.4:
+            score += 30  # 极度缩量，供应枯竭
+        elif shrink_ratio < 0.6:
+            score += 22
+        elif shrink_ratio < 0.8:
+            score += 12
+        elif shrink_ratio > 1.5:
+            score += 25  # 放量，需求进入
+        elif shrink_ratio > 1.2:
+            score += 15
+
+    # 量价配合 (20)
+    if _check_bullish_volume_divergence(df):
+        score += 20
+
+    return min(score, 100)
+
+
+def analyze_volume_pattern(df: pd.DataFrame) -> dict:
+    """量价模式分析
+
+    Returns:
+        {
+            "demand_supply_ratio": float,  # 上涨日均量/下跌日均量
+            "volume_trend": str,           # expanding/contracting/stable
+            "accumulation_score": float,   # 0-100
+        }
+    """
+    default = {"demand_supply_ratio": 1.0, "volume_trend": "stable", "accumulation_score": 0}
+
+    if len(df) < 10:
+        return default
+
+    recent = df.tail(10)
+
+    # 需求/供给比
+    up_days = recent[recent["pct_chg"] > 0]
+    down_days = recent[recent["pct_chg"] <= 0]
+    avg_up_vol = up_days["vol"].mean() if len(up_days) > 0 else 0
+    avg_down_vol = down_days["vol"].mean() if len(down_days) > 0 else 1
+    ratio = round(avg_up_vol / max(avg_down_vol, 1), 2)
+
+    # 量能趋势
+    if len(df) >= 10:
+        vol_5d_ago = df["vol"].iloc[-10:-5].mean()
+        vol_recent = df["vol"].tail(5).mean()
+        if vol_5d_ago > 0:
+            trend_ratio = vol_recent / vol_5d_ago
+            trend = "expanding" if trend_ratio > 1.2 else ("contracting" if trend_ratio < 0.8 else "stable")
+        else:
+            trend = "stable"
+    else:
+        trend = "stable"
+
+    # 累积评分
+    accumulation = 0
+    if ratio > 1.5:
+        accumulation += 40
+    if trend == "expanding" and df.iloc[-1]["pct_chg"] > 0:
+        accumulation += 30
+    if _check_volume_dry_up_before_rally(df):
+        accumulation += 30
+
+    return {
+        "demand_supply_ratio": ratio,
+        "volume_trend": trend,
+        "accumulation_score": accumulation,
+    }
+
+
+# ── 辅助函数 ──
+
+
+def _check_higher_highs(df: pd.DataFrame) -> bool:
+    """检查最近是否形成更高高点形态
+
+    取近 20 日的局部高点，检查是否有上升趋势。
+    """
+    if len(df) < 20:
+        return False
+
+    recent = df.tail(20)
+    highs = recent["high"].values
+
+    # 简单检查：最近的高点是否比之前的高点更高
+    # 将 20 日分为两段，各取最高点
+    first_half_max = highs[:10].max()
+    second_half_max = highs[10:].max()
+
+    return second_half_max >= first_half_max
+
+
+def _check_bullish_volume_divergence(df: pd.DataFrame) -> bool:
+    """检查量价配合是否健康（上涨放量，下跌缩量）"""
+    if len(df) < 6:
+        return False
+
+    recent = df.tail(5)
+    up_days = recent[recent["pct_chg"] > 0]
+    down_days = recent[recent["pct_chg"] <= 0]
+
+    if len(up_days) == 0 or len(down_days) == 0:
+        return len(up_days) > len(down_days)  # 主要是上涨日也算
+
+    avg_up_vol = up_days["vol"].mean()
+    avg_down_vol = down_days["vol"].mean()
+
+    return avg_up_vol > avg_down_vol
+
+
+def _check_volume_dry_up_before_rally(df: pd.DataFrame) -> bool:
+    """检查是否有缩量后放量的突破前奏
+
+    经典突破模式：量能萎缩至极低，然后在突破日放量。
+    """
+    if len(df) < 15:
+        return False
+
+    last = df.iloc[-1]
+
+    # 检查近 5-10 日内是否有极度缩量日
+    vol_ma_col = "vol_ma10" if "vol_ma10" in df.columns else "vol_ma5"
+    if vol_ma_col not in df.columns:
+        return False
+
+    recent_10 = df.tail(10)
+    had_dry_up = any(
+        row["vol"] < row[vol_ma_col] * 0.5
+        for _, row in recent_10.iterrows()
+        if not pd.isna(row[vol_ma_col]) and row[vol_ma_col] > 0
+    )
+
+    # 且当日量能放大
+    vol_ma_col_5 = "vol_ma5" if "vol_ma5" in df.columns else vol_ma_col
+    vol_expanding = (
+        not pd.isna(last[vol_ma_col_5])
+        and last[vol_ma_col_5] > 0
+        and last["vol"] > last[vol_ma_col_5] * 1.2
+    )
+
+    return had_dry_up and vol_expanding

backend/app/analysis/intraday.py

@@ -266,3 +266,64 @@ def _score_intraday(quote: StockQuote) -> float:
             score += 3
 
     return score
+
+
+async def intraday_sector_scan(prev_sectors: list[SectorInfo]) -> list[SectorInfo]:
+    """盘中板块热度更新：用腾讯实时行情刷新板块涨幅和涨停数
+
+    基于前一日的板块列表（来自 Tushare），用成分股的实时行情
+    重新计算板块涨跌幅和涨停家数。
+    """
+    if not prev_sectors:
+        return prev_sectors
+
+    # 收集所有板块的成分股
+    sector_members: dict[str, list[str]] = {}
+    all_codes = set()
+    for sector in prev_sectors:
+        members = tushare_client.get_ths_members(sector.sector_code)
+        if members.empty or "con_code" not in members.columns:
+            continue
+        codes = [c for c in members["con_code"].tolist() if "." in str(c)]
+        sector_members[sector.sector_code] = codes
+        all_codes.update(codes)
+
+    if not all_codes:
+        return prev_sectors
+
+    # 批量获取实时行情
+    quotes = await tencent_client.get_realtime_quotes_batch(list(all_codes))
+    if not quotes:
+        return prev_sectors
+
+    # 构建涨停集合
+    limit_up_codes = set()
+    for code, q in quotes.items():
+        if q.limit_up and q.price >= q.limit_up * 0.995:
+            limit_up_codes.add(code)
+
+    # 更新每个板块的数据
+    for sector in prev_sectors:
+        codes = sector_members.get(sector.sector_code, [])
+        if not codes:
+            continue
+
+        sector_quotes = [quotes[c] for c in codes if c in quotes]
+        if not sector_quotes:
+            continue
+
+        # 实时涨跌幅（成分股均值）
+        pct_changes = [q.pct_chg for q in sector_quotes if q.pct_chg is not None]
+        if pct_changes:
+            sector.pct_change = round(sum(pct_changes) / len(pct_changes), 2)
+
+        # 实时涨停家数
+        sector.limit_up_count = len([c for c in codes if c in limit_up_codes])
+
+    logger.info(
+        f"盘中板块实时更新: {len(prev_sectors)} 个板块, "
+        f"涨幅最高={max(prev_sectors, key=lambda s: s.pct_change).sector_name} "
+        f"({max(s.pct_change for s in prev_sectors):.1f}%)"
+    )
+
+    return prev_sectors

backend/app/analysis/sector_scanner.py

@@ -1,11 +1,17 @@
 """板块热度扫描
 
 综合板块涨幅、资金净流入、涨停家数、持续性，
-输出热门板块排名。
+输出热门板块排名及深度分析。
 
 优化策略：先用板块资金流向批量数据预筛 Top 板块，
 只对 Top 板块做逐个详细查询（ths_daily/ths_member），
 避免遍历全部数百个板块导致大量 API 调用。
+
+增强分析：
+  - 领涨股：每个板块中涨幅前3的成分股
+  - 资金趋势：近5日板块资金净流入走势
+  - 涨跌趋势：近5日板块涨跌幅走势
+  - 主力资金占比
 """
 
 import logging
@@ -37,7 +43,7 @@ def _normalize_score(values: list[float], reverse: bool = False) -> list[float]:
 
 
 def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
-    """扫描热门板块，返回按热度排名的板块列表"""
+    """扫描热门板块，返回按热度排名的板块列表（含深度分析）"""
     if not trade_date:
         trade_date = tushare_client.get_latest_trade_date()
 
@@ -54,18 +60,27 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
     logger.info(f"板块资金流向预筛: {len(sector_mf)} 个板块 -> Top {len(top_codes)} 进入详细分析")
 
     # 构建资金流向查找表
+    # Tushare moneyflow_ind_ths 的金额单位是亿元，统一转换为万元
+    _UNIT_CONV = 10000
     mf_lookup = {}
+    # 同时构建主力买卖数据（用于计算主力占比）
+    mf_detail = {}
     for _, row in sector_mf.iterrows():
-        mf_lookup[row["ts_code"]] = float(row["net_amount"])
+        mf_lookup[row["ts_code"]] = float(row["net_amount"]) * _UNIT_CONV
+        mf_detail[row["ts_code"]] = {
+            "net_amount": float(row["net_amount"]) * _UNIT_CONV,
+            "buy_elg_amount": float(row.get("buy_elg_amount", 0)) * _UNIT_CONV if pd.notna(row.get("buy_elg_amount")) else 0,
+            "sell_elg_amount": float(row.get("sell_elg_amount", 0)) * _UNIT_CONV if pd.notna(row.get("sell_elg_amount")) else 0,
+            "buy_lg_amount": float(row.get("buy_lg_amount", 0)) * _UNIT_CONV if pd.notna(row.get("buy_lg_amount")) else 0,
+            "sell_lg_amount": float(row.get("sell_lg_amount", 0)) * _UNIT_CONV if pd.notna(row.get("sell_lg_amount")) else 0,
+        }
 
     # 构建板块名称查找表
-    # moneyflow_ind_ths 返回的是行业板块（881XXX.TI），自带 industry 列
     name_lookup = {}
     if "industry" in sector_mf.columns:
         for _, r in sector_mf.iterrows():
             if pd.notna(r.get("industry")):
                 name_lookup[r["ts_code"]] = str(r["industry"])
-    # 补充：从 ths_index type=I（行业板块）获取名称
     index_list = tushare_client.get_ths_index_list("I")
     if not index_list.empty:
         for _, r in index_list.iterrows():
@@ -75,22 +90,34 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
     # ── 第二步：获取涨跌停列表（1 次 API 调用）──
     limit_df = tushare_client.get_limit_list(trade_date)
     limit_up_codes = set()
+    # 同时收集涨停股的涨跌幅信息（用于领涨股展示）
+    limit_up_info: dict[str, dict] = {}
     if not limit_df.empty:
         up_df = limit_df[limit_df["limit"] == "U"]
         up_df = up_df[~up_df["name"].str.contains("ST", na=False)]
         limit_up_codes = set(up_df["ts_code"].tolist())
+        for _, row in up_df.iterrows():
+            limit_up_info[row["ts_code"]] = {
+                "name": row["name"],
+                "pct_chg": float(row.get("pct_chg", 10)),
+                "limit_times": int(row.get("limit_times", 1)),
+            }
 
-    # ── 第三步：只对 Top 板块做逐个详细查询 ──
+    # ── 第三步：获取全市场日线数据（用于领涨股计算）──
+    daily_all = tushare_client.get_daily_all(trade_date)
+    stock_basic = tushare_client.get_stock_basic()
+
+    # ── 第四步：只对 Top 板块做逐个详细查询 ──
     sectors = []
 
     for ts_code in top_codes:
-        # 板块名称从 ths_index 查找表获取
         sector_name = name_lookup.get(ts_code, ts_code)
 
-        # 板块日线 - 获取近5日数据（1 次 API）
+        # 板块日线 - 获取近5日数据
         ths_daily = tushare_client.get_ths_daily(ts_code, days=5)
         pct_change = 0.0
         days_continuous = 0
+        pct_trend: list[float] = []
 
         if not ths_daily.empty:
             ths_daily = ths_daily.sort_values("trade_date")
@@ -101,6 +128,9 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
                 today_data = ths_daily.tail(1)
             pct_change = float(today_data["pct_change"].iloc[0]) if not today_data.empty else 0
 
+            # 近5日涨跌幅趋势
+            pct_trend = [round(float(d["pct_change"]), 2) for _, d in ths_daily.iterrows()]
+
             # 连续上涨天数
             for _, d in ths_daily.iloc[::-1].iterrows():
                 if d["pct_change"] > 0:
@@ -108,15 +138,75 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
                 else:
                     break
 
-        # 板块资金净流入（从预筛数据中直接取）
+        # 板块资金净流入
         capital_inflow = mf_lookup.get(ts_code, 0.0)
 
-        # 板块内涨停家数（1 次 API）
+        # 主力资金占比 = (特大单净买 + 大单净买) / (特大单买卖总额 + 大单买卖总额)
+        main_force_ratio = 0.0
+        detail = mf_detail.get(ts_code, {})
+        total_main_amount = (detail.get("buy_elg_amount", 0) + detail.get("sell_elg_amount", 0) +
+                             detail.get("buy_lg_amount", 0) + detail.get("sell_lg_amount", 0))
+        if total_main_amount > 0:
+            main_force_ratio = round(capital_inflow / total_main_amount * 100, 1)
+
+        # 板块成分股分析
         limit_up_count = 0
+        leading_stocks: list[dict] = []
+        member_count = 0
+        turnover_avg = 0.0
+
         members = tushare_client.get_ths_members(ts_code)
         if not members.empty and "con_code" in members.columns:
-            member_codes = set(members["con_code"].tolist())
-            limit_up_count = len(limit_up_codes & member_codes)
+            member_codes = list(members["con_code"].tolist())
+            member_set = set(member_codes)
+            member_count = len(member_codes)
+            limit_up_count = len(limit_up_codes & member_set)
+
+            # 领涨股：从当日全市场日级数据中筛选该板块成分股，按涨幅排序取前3
+            if not daily_all.empty:
+                sector_daily = daily_all[daily_all["ts_code"].isin(member_set)].copy()
+                # 排除 ST
+                if not stock_basic.empty:
+                    st_set = set(stock_basic[stock_basic["name"].str.contains("ST", na=False)]["ts_code"])
+                    sector_daily = sector_daily[~sector_daily["ts_code"].isin(st_set)]
+
+                sector_daily = sector_daily.sort_values("pct_chg", ascending=False)
+                # 计算板块平均换手率
+                if "turnover_rate" in sector_daily.columns:
+                    turnover_values = sector_daily["turnover_rate"].dropna()
+                    if len(turnover_values) > 0:
+                        turnover_avg = round(float(turnover_values.mean()), 2)
+
+                # 构建名称查找
+                name_map = {}
+                if not stock_basic.empty:
+                    for _, br in stock_basic.iterrows():
+                        name_map[br["ts_code"]] = br["name"]
+                if "con_name" in members.columns:
+                    for _, m in members.iterrows():
+                        if pd.notna(m.get("con_name")):
+                            name_map[m["con_code"]] = m["con_name"]
+
+                # 取涨幅前3
+                for _, sr in sector_daily.head(3).iterrows():
+                    leading_stocks.append({
+                        "ts_code": sr["ts_code"],
+                        "name": name_map.get(sr["ts_code"], sr["ts_code"]),
+                        "pct_chg": round(float(sr["pct_chg"]), 2),
+                        "amount": round(float(sr.get("amount", 0)), 0),
+                    })
+
+            # 涨停股也在成分股中的，补充到领涨股（如未在top3中）
+            for code in (limit_up_codes & member_set):
+                if code not in [s["ts_code"] for s in leading_stocks]:
+                    info = limit_up_info.get(code, {})
+                    leading_stocks.append({
+                        "ts_code": code,
+                        "name": info.get("name", code),
+                        "pct_chg": info.get("pct_chg", 10),
+                        "amount": 0,
+                        "limit_times": info.get("limit_times", 1),
+                    })
 
         sectors.append(SectorInfo(
             sector_code=ts_code,
@@ -125,6 +215,11 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
             capital_inflow=round(capital_inflow, 2),
             limit_up_count=limit_up_count,
             days_continuous=days_continuous,
+            member_count=member_count,
+            leading_stocks=leading_stocks,
+            pct_trend=pct_trend,
+            turnover_avg=turnover_avg,
+            main_force_ratio=main_force_ratio,
         ))
 
     if not sectors:
@@ -133,13 +228,13 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
     # ── 板块阶段判定 ──
     for s in sectors:
         if s.days_continuous <= 2:
-            s.stage = "early"    # 启动期，安全
+            s.stage = "early"
         elif s.days_continuous == 3:
-            s.stage = "mid"      # 发展期，正常
+            s.stage = "mid"
         elif s.days_continuous == 4:
-            s.stage = "late"     # 后期，谨慎
+            s.stage = "late"
         else:
-            s.stage = "end"      # 尾声，高风险
+            s.stage = "end"
 
     # ── 综合评分 ──
     pct_scores = _normalize_score([s.pct_change for s in sectors])
@@ -154,20 +249,17 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
             lim_scores[i] * 0.25 +
             con_scores[i] * 0.20
         )
-        # 连续2日以上资金流入的板块加分
         if s.days_continuous >= 2:
             heat += 5
-        # 首日大涨（昨日涨幅<=0，今日>3%）可视为新热点，加分
         if s.days_continuous == 1 and s.pct_change > 3:
             heat += 3
 
-        # 板块阶段调整：早期加分，尾声减分（防追高）
         if s.stage == "early":
-            heat += 8   # 启动期，介入安全，大幅加分
+            heat += 8
         elif s.stage == "late":
-            heat -= 5   # 后期，风险上升
+            heat -= 5
         elif s.stage == "end":
-            heat -= 12  # 尾声，大幅减分
+            heat -= 12
 
         s.heat_score = round(max(0, min(heat, 100)), 1)
 
@@ -175,7 +267,10 @@ def scan_hot_sectors(trade_date: str = None) -> list[SectorInfo]:
 
     top = sectors[:settings.top_sector_count]
     for s in top:
-        logger.info(f"热门板块: {s.sector_name} 涨幅{s.pct_change}% 资金{s.capital_inflow}万 "
-                     f"涨停{s.limit_up_count} 连续{s.days_continuous}天 阶段={s.stage} 热度{s.heat_score}")
+        leaders = ", ".join(f'{l["name"]}({l["pct_chg"]}%)' for l in s.leading_stocks[:3])
+        inflow_display = f"{s.capital_inflow / 10000:.1f}亿" if s.capital_inflow >= 10000 else f"{s.capital_inflow:.0f}万"
+        logger.info(f"热门板块: {s.sector_name} 涨幅{s.pct_change}% 资金{inflow_display} "
+                     f"涨停{s.limit_up_count} 连续{s.days_continuous}天 阶段={s.stage} 热度{s.heat_score} "
+                     f"领涨=[{leaders}]")
 
     return sectors

backend/app/analysis/technical.py

@@ -83,6 +83,7 @@ def add_all_indicators(df: pd.DataFrame) -> pd.DataFrame:
     # 量均线
     df["vol_ma5"] = calc_volume_ma(vol, 5)
     df["vol_ma10"] = calc_volume_ma(vol, 10)
+    df["vol_ma20"] = calc_volume_ma(vol, 20)
 
     # 涨跌幅（如果没有 pct_chg 列）
     if "pct_chg" not in df.columns:

backend/app/analysis/trend_scanner.py

@@ -0,0 +1,516 @@
+"""趋势突破三阶段扫描管道
+
+Phase 1: 全市场批量预筛 — get_daily_all × 5 天 + get_daily_basic → ~300 候选
+Phase 2: 资金流批量过滤 — get_moneyflow_batch → ~80 候选
+Phase 3: 逐股深度分析 — get_stock_daily(120d) + 入场信号 + 供需评分 → ~20 推荐
+"""
+
+import logging
+from datetime import datetime, timedelta
+
+import numpy as np
+import pandas as pd
+
+from app.config import settings
+from app.data.tushare_client import tushare_client
+from app.analysis.technical import add_all_indicators
+from app.analysis.breakout_signals import (
+    classify_entry_signal,
+    score_trend_timing,
+    score_supply_demand,
+    analyze_volume_pattern,
+    EntrySignal,
+)
+
+logger = logging.getLogger(__name__)
+
+
+async def scan_trend_breakout(
+    trade_date: str = None,
+    market_temp=None,
+    hot_sectors: list = None,
+    intraday: bool = False,
+) -> list[dict]:
+    """统一趋势突破扫描 — 三阶段管道
+
+    Returns:
+        list[dict] — 每个字典包含:
+            ts_code, name, sector, entry_signal_type, entry_signal_score,
+            trend_timing_score, supply_demand_score, capital_score,
+            main_net_inflow, inflow_ratio, turnover_rate, volume_ratio,
+            circ_mv, pe, pb, volume_trend, demand_supply_ratio
+    """
+    if not trade_date:
+        trade_date = tushare_client.get_latest_trade_date()
+
+    logger.info(f"=== 趋势突破扫描 (trade_date={trade_date}) ===")
+
+    # Phase 1: 批量预筛
+    candidates = _bulk_pre_filter(trade_date)
+    if candidates.empty:
+        logger.info("Phase 1 预筛无候选")
+        return []
+
+    logger.info(f"Phase 1 预筛: {len(candidates)} 只候选")
+
+    # Phase 2: 资金流过滤
+    candidates = _bulk_capital_filter(candidates, trade_date)
+    if candidates.empty:
+        logger.info("Phase 2 资金过滤后无候选")
+        return []
+
+    logger.info(f"Phase 2 资金过滤: {len(candidates)} 只候选")
+
+    # Phase 3: 逐股深度分析
+    results = await _deep_analysis(candidates, trade_date, market_temp, hot_sectors or [], intraday)
+
+    logger.info(f"Phase 3 深度分析: {len(results)} 只推荐")
+    return results
+
+
+def _bulk_pre_filter(trade_date: str) -> pd.DataFrame:
+    """Phase 1: 批量预筛
+
+    利用 get_daily_all × 5 天构建迷你时序，用 5 日趋势初筛。
+    合并 get_daily_basic 过滤市值/换手率。
+    排除 ST 和次新股。
+    """
+    # 获取交易日历
+    dates = tushare_client.get_trade_dates()
+    if not dates:
+        logger.warning("Phase 1: 无法获取交易日历")
+        return pd.DataFrame()
+
+    # 取最近 5 个交易日（含当日）
+    recent_dates = dates[-5:] if len(dates) >= 5 else dates
+    if trade_date not in recent_dates:
+        recent_dates = recent_dates[-4:] + [trade_date] if len(recent_dates) >= 4 else [trade_date]
+
+    logger.info(f"Phase 1: 使用交易日 {recent_dates}")
+
+    # 获取多日全市场数据
+    daily_frames = []
+    for d in recent_dates:
+        df = tushare_client.get_daily_all(d)
+        if not df.empty:
+            daily_frames.append(df)
+            logger.debug(f"Phase 1: {d} 获取到 {len(df)} 只股票数据")
+        else:
+            logger.warning(f"Phase 1: {d} 无数据")
+
+    if len(daily_frames) < 2:
+        # 至少需要 2 天数据做趋势判断
+        if daily_frames:
+            logger.info("Phase 1: 仅有1天数据，跳过趋势筛选")
+            return _filter_daily_basic(daily_frames[0], trade_date)
+        logger.warning("Phase 1: 完全无数据")
+        return pd.DataFrame()
+
+    # 拼接多日数据
+    all_daily = pd.concat(daily_frames, ignore_index=True)
+    logger.info(f"Phase 1: 拼接后总记录数 {len(all_daily)}")
+
+    # 计算每只股票的 5 日趋势
+    stock_groups = all_daily.groupby("ts_code")
+
+    # 分步统计
+    total_stocks = len(stock_groups)
+    count_has_data = 0
+    count_main_board = 0
+    count_no_limit = 0
+    count_positive_return = 0
+    count_above_avg = 0
+    count_near_high = 0
+
+    trend_data = []
+    for ts_code, group in stock_groups:
+        if len(group) < 2:
+            continue
+        count_has_data += 1
+
+        group = group.sort_values("trade_date")
+        latest = group.iloc[-1]
+
+        # 排除非主板（只保留 00/30/60 开头）
+        code_prefix = ts_code[:2]
+        if code_prefix not in ("00", "30", "60"):
+            continue
+        count_main_board += 1
+
+        # 排除涨跌停
+        pct = latest["pct_chg"]
+        if pd.isna(pct) or abs(pct) > 9.5:
+            continue
+        count_no_limit += 1
+
+        # 5 日涨幅
+        first_close = group.iloc[0]["close"]
+        last_close = latest["close"]
+        if first_close <= 0:
+            continue
+        ret_5d = (last_close - first_close) / first_close * 100
+
+        # 条件：5 日涨幅 > -2%（允许小幅回调） 且 < 20%（未过热）
+        if ret_5d <= -2 or ret_5d >= 20:
+            continue
+        count_positive_return += 1
+
+        # 5 日均价
+        avg_close = group["close"].mean()
+        if avg_close <= 0:
+            continue
+
+        # 收盘价 >= 5 日均价 * 0.98（允许略低于均线）
+        if last_close < avg_close * 0.98:
+            continue
+        count_above_avg += 1
+
+        # 距 5 日高点 < 8%（放宽，不要求紧贴高点）
+        high_5d = group["high"].max()
+        if high_5d <= 0:
+            continue
+        dist_from_high = (high_5d - last_close) / high_5d * 100
+        if dist_from_high > 8:
+            continue
+        count_near_high += 1
+
+        trend_data.append({
+            "ts_code": ts_code,
+            "close": last_close,
+            "pct_chg_today": pct,
+            "return_5d": round(ret_5d, 2),
+            "dist_from_high_5d": round(dist_from_high, 2),
+            "vol_today": latest["vol"],
+        })
+
+    logger.info(
+        f"Phase 1 趋势筛选: "
+        f"总{total_stocks} → 有数据{count_has_data} → 主板{count_main_board} → "
+        f"非涨跌停{count_no_limit} → 5日涨幅>-2%{count_positive_return} → "
+        f"近均线{count_above_avg} → 近高点{count_near_high}"
+    )
+
+    if not trend_data:
+        return pd.DataFrame()
+
+    candidates = pd.DataFrame(trend_data)
+
+    # 合并 daily_basic 过滤（使用更宽松的换手率）
+    return _filter_daily_basic(candidates, trade_date)
+
+
+def _filter_daily_basic(candidates: pd.DataFrame, trade_date: str) -> pd.DataFrame:
+    """使用 daily_basic 过滤市值、换手率，排除 ST 和次新"""
+    basic = tushare_client.get_daily_basic(trade_date)
+    if basic.empty:
+        logger.warning("Phase 1: daily_basic 无数据")
+        return candidates
+
+    # 股票基本信息（排除 ST 和次新）
+    stock_basic = tushare_client.get_stock_basic()
+    exclude_codes = set()
+    if not stock_basic.empty:
+        # ST
+        st_codes = set(stock_basic[stock_basic["name"].str.contains("ST", na=False)]["ts_code"])
+        exclude_codes.update(st_codes)
+        # 次新
+        cutoff = (datetime.now() - timedelta(days=settings.min_list_days)).strftime("%Y%m%d")
+        new_codes = set(stock_basic[stock_basic["list_date"] > cutoff]["ts_code"])
+        exclude_codes.update(new_codes)
+
+    # Tushare circ_mv 单位是万元，转为亿元
+    basic["circ_mv"] = basic["circ_mv"] / 10000
+
+    # Phase 1 使用更宽松的换手率门槛（2%）以保留更多候选
+    min_tr = min(settings.min_turnover_rate, 2.0)
+    basic_filtered = basic[
+        (basic["circ_mv"] >= settings.min_circ_mv) &
+        (basic["circ_mv"] <= settings.max_circ_mv) &
+        (basic["turnover_rate"] >= min_tr) &
+        (basic["turnover_rate"] <= settings.max_turnover_rate) &
+        (~basic["ts_code"].isin(exclude_codes))
+    ].copy()
+
+    logger.info(
+        f"Phase 1 daily_basic: 全市场{len(basic)}只 → "
+        f"过滤后{len(basic_filtered)}只 "
+        f"(circ_mv {settings.min_circ_mv}-{settings.max_circ_mv}亿, "
+        f"turnover >={min_tr}%)"
+    )
+
+    if basic_filtered.empty:
+        return pd.DataFrame()
+
+    # 合并候选和 basic 数据
+    if "circ_mv" in candidates.columns:
+        merged = candidates
+    else:
+        # candidates 没有 basic 列，通过 ts_code 合并
+        basic_subset = basic_filtered[["ts_code", "turnover_rate", "volume_ratio",
+                                       "circ_mv", "pe", "pb"]].drop_duplicates("ts_code")
+        # 如果 candidates 有 ts_code 列
+        if "ts_code" in candidates.columns:
+            merged = candidates.merge(basic_subset, on="ts_code", how="inner")
+        else:
+            # candidates 是 get_daily_all 返回的原始 df
+            merged = basic_subset
+
+    logger.info(f"Phase 1 合并后: {len(merged)} 只候选")
+    return merged
+
+
+def _bulk_capital_filter(candidates: pd.DataFrame, trade_date: str) -> pd.DataFrame:
+    """Phase 2: 资金流批量过滤
+
+    硬条件：主力净流入 > 0（特大单 + 大单净买为正）
+    """
+    mf = tushare_client.get_moneyflow_batch(trade_date)
+    if mf.empty:
+        return candidates
+
+    # 计算主力净流入 = (特大单买入-特大单卖出) + (大单买入-大单卖出)
+    mf["main_net_inflow"] = (
+        (mf["buy_elg_amount"] - mf["sell_elg_amount"]) +
+        (mf["buy_lg_amount"] - mf["sell_lg_amount"])
+    )
+
+    # 计算流入比例
+    total = (
+        mf["buy_elg_amount"] + mf["sell_elg_amount"] +
+        mf["buy_lg_amount"] + mf["sell_lg_amount"] +
+        mf["buy_md_amount"] + mf["sell_md_amount"] +
+        mf["buy_sm_amount"] + mf["sell_sm_amount"]
+    )
+    mf["inflow_ratio"] = mf["main_net_inflow"] / total.replace(0, np.nan) * 100
+    mf["inflow_ratio"] = mf["inflow_ratio"].fillna(0)
+
+    # 只保留主力净流入 > 0 的
+    mf_positive = mf[mf["main_net_inflow"] > 0][["ts_code", "main_net_inflow", "inflow_ratio"]]
+
+    # 合并候选
+    if "ts_code" in candidates.columns:
+        merged = candidates.merge(mf_positive, on="ts_code", how="inner")
+    else:
+        merged = mf_positive
+
+    # 按主力净流入排序，取 top 100
+    if not merged.empty:
+        merged = merged.sort_values("main_net_inflow", ascending=False).head(100)
+
+    return merged.reset_index(drop=True)
+
+
+async def _deep_analysis(
+    candidates: pd.DataFrame,
+    trade_date: str,
+    market_temp,
+    hot_sectors: list,
+    intraday: bool,
+) -> list[dict]:
+    """Phase 3: 逐股深度分析
+
+    对每只候选获取 120 日 K 线，计算技术指标，
+    分类入场信号，评分供需关系。
+    """
+    import asyncio
+    from app.analysis.signals import generate_signals
+    from app.analysis.capital_flow import _score_valuation
+
+    # 获取股票名称映射
+    stock_basic = tushare_client.get_stock_basic()
+    name_map = {}
+    industry_map = {}
+    if not stock_basic.empty:
+        for _, row in stock_basic.iterrows():
+            name_map[row["ts_code"]] = row["name"]
+            industry_map[row["ts_code"]] = row.get("industry", "")
+
+    # 获取多日资金流（用于资金持续性评分）
+    mf_history = _get_multi_day_moneyflow(candidates, trade_date)
+
+    results = []
+    total = len(candidates)
+    signal_counts = {"breakout": 0, "pullback": 0, "launch": 0, "none": 0}
+
+    for idx, row in candidates.iterrows():
+        ts_code = row["ts_code"] if "ts_code" in candidates.columns else ""
+
+        if not ts_code:
+            continue
+
+        name = name_map.get(ts_code, ts_code)
+        sector = industry_map.get(ts_code, "")
+
+        try:
+            # 获取 120 日 K 线
+            df = tushare_client.get_stock_daily(ts_code, 120)
+            if df.empty or len(df) < 30:
+                continue
+
+            # 添加技术指标
+            df = add_all_indicators(df)
+
+            # 入场信号分类
+            entry_signal = classify_entry_signal(df)
+            signal_type = entry_signal["signal_type"]
+            if signal_type == EntrySignal.NONE:
+                signal_counts["none"] += 1
+                continue
+            signal_counts[signal_type.value] += 1
+
+            # 趋势&时机评分
+            trend_score = score_trend_timing(df, entry_signal)
+
+            # 供需评分
+            sd_score = score_supply_demand(df)
+
+            # 量价模式分析
+            vol_pattern = analyze_volume_pattern(df)
+
+            # 技术信号（复用现有 generate_signals）
+            tech_signal = generate_signals(ts_code, name)
+
+            # 资金流评分
+            capital_score = _score_capital(
+                row, mf_history.get(ts_code, pd.DataFrame())
+            )
+
+            # 估值评分（作为辅助参考）
+            pe = row.get("pe")
+            pb = row.get("pb")
+            valuation_score = _score_valuation(pe, pb)
+
+            results.append({
+                "ts_code": ts_code,
+                "name": name,
+                "sector": sector,
+                "entry_signal_type": entry_signal["signal_type"].value,
+                "entry_signal_score": round(entry_signal["signal_score"], 1),
+                "entry_signal_details": entry_signal.get("details", {}),
+                "trend_timing_score": round(trend_score, 1),
+                "supply_demand_score": round(sd_score, 1),
+                "capital_score": round(capital_score, 1),
+                "valuation_score": round(valuation_score, 1),
+                "technical_score": round(tech_signal.score, 1),
+                "position_score": round(tech_signal.position_score, 1),
+                "main_net_inflow": row.get("main_net_inflow", 0),
+                "inflow_ratio": round(row.get("inflow_ratio", 0), 2),
+                "turnover_rate": row.get("turnover_rate"),
+                "volume_ratio": row.get("volume_ratio"),
+                "circ_mv": row.get("circ_mv"),
+                "pe": pe,
+                "pb": pb,
+                "volume_trend": vol_pattern["volume_trend"],
+                "demand_supply_ratio": vol_pattern["demand_supply_ratio"],
+                # 技术信号详情（用于生成推荐理由）
+                "tech_signal": tech_signal,
+            })
+
+            if len(results) >= settings.top_stock_count:
+                break
+
+        except Exception as e:
+            logger.debug(f"深度分析 {ts_code} 失败: {e}")
+            continue
+
+        # 让出控制权，避免阻塞事件循环
+        if idx % 10 == 0:
+            await asyncio.sleep(0)
+
+    logger.info(
+        f"Phase 3 入场信号分布: "
+        f"突破={signal_counts['breakout']} 回踩={signal_counts['pullback']} "
+        f"启动={signal_counts['launch']} 无信号={signal_counts['none']} "
+        f"(共分析{total}只)"
+    )
+
+    return results
+
+
+def _get_multi_day_moneyflow(candidates: pd.DataFrame, trade_date: str) -> dict[str, pd.DataFrame]:
+    """获取候选股票近 5 日资金流数据（用于资金持续性评分）
+
+    利用 get_stock_moneyflow 逐只获取（仅对 Phase 2 过滤后的 ~80 只）。
+    """
+    result = {}
+
+    if "ts_code" not in candidates.columns:
+        return result
+
+    for ts_code in candidates["ts_code"].values[:80]:
+        try:
+            df = tushare_client.get_stock_moneyflow(ts_code, 5)
+            if not df.empty:
+                result[ts_code] = df
+        except Exception:
+            pass
+
+    return result
+
+
+def _score_capital(stock_row: dict | pd.Series, mf_history: pd.DataFrame) -> float:
+    """资金流评分 (0-100)
+
+    维度：
+    - 当日主力净流入规模 (35)
+    - 资金持续性 (35): 近 N 日中主力净流入为正的天数
+    - 流入比 (15)
+    - 量比 (15)
+    """
+    score = 0
+
+    main_net = stock_row.get("main_net_inflow", 0) or 0
+    inflow_ratio = stock_row.get("inflow_ratio", 0) or 0
+    volume_ratio = stock_row.get("volume_ratio")
+
+    # 当日主力净流入规模 (35)
+    if main_net > 10000:
+        score += 35
+    elif main_net > 5000:
+        score += 28
+    elif main_net > 2000:
+        score += 20
+    elif main_net > 500:
+        score += 12
+    elif main_net > 0:
+        score += 5
+
+    # 资金持续性 (35)
+    if not mf_history.empty and len(mf_history) >= 2:
+        positive_days = 0
+        for _, r in mf_history.iterrows():
+            net = (r.get("buy_elg_amount", 0) - r.get("sell_elg_amount", 0) +
+                   r.get("buy_lg_amount", 0) - r.get("sell_lg_amount", 0))
+            if net > 0:
+                positive_days += 1
+        total_days = len(mf_history)
+        if total_days >= 3 and positive_days >= total_days * 0.7:
+            score += 35
+        elif total_days >= 2 and positive_days >= total_days * 0.5:
+            score += 25
+        elif positive_days >= 1:
+            score += 12
+    else:
+        # 只有一天数据且为正（已经通过 Phase 2 过滤）
+        score += 15
+
+    # 流入比 (15)
+    if inflow_ratio > 15:
+        score += 15
+    elif inflow_ratio > 10:
+        score += 12
+    elif inflow_ratio > 5:
+        score += 8
+    elif inflow_ratio > 0:
+        score += 4
+
+    # 量比 (15)
+    if volume_ratio:
+        if volume_ratio > 2.0:
+            score += 15
+        elif volume_ratio > 1.5:
+            score += 10
+        elif volume_ratio > 1.0:
+            score += 5
+
+    return min(score, 100)

backend/app/api/recommendations.py

@@ -52,6 +52,7 @@ async def get_latest():
                 "llm_analysis": r.llm_analysis,
                 "llm_score": r.llm_score,
                 "strategy": r.strategy,
+                "entry_signal_type": r.entry_signal_type,
                 "scan_session": r.scan_session,
                 "created_at": r.created_at.isoformat() if r.created_at else None,
             }
@@ -88,6 +89,5 @@ async def get_scan_status():
 
 @router.get("/history")
 async def get_history(days: int = 7):
-    """获取历史推荐"""
-    rows = await get_recommendation_history(days)
-    return rows
+    """获取历史推荐（按日期分组）"""
+    return await get_recommendation_history(days)

backend/app/api/sectors.py

@@ -66,13 +66,26 @@ async def _enrich_sectors_realtime(sectors_data: list[dict]) -> list[dict]:
                 1 for q in member_quotes
                 if q.limit_up and q.price >= q.limit_up * 0.995
             )
+
+            # 盘中更新领涨股
+            sorted_quotes = sorted(member_quotes, key=lambda q: q.pct_chg, reverse=True)
+            s["leading_stocks_realtime"] = [
+                {
+                    "ts_code": q.ts_code,
+                    "name": q.name or q.ts_code,
+                    "pct_chg": round(q.pct_chg, 2),
+                    "amount": round(q.amount, 0),
+                }
+                for q in sorted_quotes[:3]
+            ]
         else:
             s["realtime_pct_change"] = None
             s["realtime_limit_up_count"] = None
+            s["leading_stocks_realtime"] = None
 
         s["is_realtime"] = True
 
-    # 盘中按实时涨幅重新排序（涨幅高的排前面）
+    # 盘中按实时涨幅重新排序
     sectors_data.sort(key=lambda s: s.get("realtime_pct_change") or 0, reverse=True)
 
     return sectors_data
@@ -92,6 +105,12 @@ async def get_hot_sectors(limit: int = 10):
             "days_continuous": s.days_continuous,
             "heat_score": s.heat_score,
             "stage": s.stage,
+            # 增强分析字段
+            "member_count": s.member_count,
+            "leading_stocks": s.leading_stocks,
+            "pct_trend": s.pct_trend,
+            "turnover_avg": s.turnover_avg,
+            "main_force_ratio": s.main_force_ratio,
         }
         for s in sectors[:limit]
     ]

backend/app/config.py

@@ -43,6 +43,11 @@ class Settings(BaseSettings):
     # 风控
     stop_loss_pct: float = 5.0          # 止损比例 %
 
+    # 趋势突破策略参数
+    breakout_min_volume_ratio: float = 1.2    # 突破型最小量比
+    pullback_max_shrink_ratio: float = 0.85   # 回踩型最大缩量比
+    consolidation_max_range_pct: float = 8.0  # 启动型最大整理振幅 %
+
     # LLM (DeepSeek)
     deepseek_api_key: str = ""
     deepseek_base_url: str = "https://api.deepseek.com/v1"

backend/app/data/models.py

@@ -57,12 +57,20 @@ class SectorInfo(BaseModel):
     sector_code: str
     sector_name: str
     pct_change: float = 0       # 涨跌幅 %
-    capital_inflow: float = 0   # 主力净流入（万）
+    capital_inflow: float = 0       # 主力净流入（万元，原始数据来自Tushare亿元×10000）
     limit_up_count: int = 0     # 涨停数
     days_continuous: int = 0    # 连续资金流入天数
     heat_score: float = 0       # 热度综合评分
     stage: str = "mid"          # 板块阶段: early/mid/late/end
 
+    # ── 板块分析增强字段 ──
+    member_count: int = 0               # 成分股数量
+    leading_stocks: list[dict] = []     # 领涨股 [{ts_code, name, pct_chg, amount}]
+    capital_trend: list[float] = []     # 近5日资金净流入趋势（万）
+    pct_trend: list[float] = []         # 近5日涨跌幅趋势
+    turnover_avg: float = 0             # 板块平均换手率
+    main_force_ratio: float = 0         # 主力资金占比（主力净流入/总成交额）
+
 
 class MarketTemperature(BaseModel):
     trade_date: str
@@ -119,7 +127,8 @@ class Recommendation(BaseModel):
     reasons: list[str] = []
     risk_note: str = ""
     level: str = ""                 # 强烈推荐/推荐/观望/回避
-    strategy: str = "momentum"      # momentum(强中选强) / potential(潜在启动)
+    strategy: str = "trend_breakout"  # trend_breakout / momentum(旧) / potential(旧)
+    entry_signal_type: str = "none"   # breakout / pullback / launch / none
     llm_analysis: str = ""          # LLM 深度分析
     llm_score: float | None = None  # AI 评分 1-10
     scan_session: str = ""

backend/app/db/database.py

@@ -33,6 +33,7 @@ async def init_db():
             "ALTER TABLE recommendations ADD COLUMN llm_score REAL",
             "ALTER TABLE market_temperature ADD COLUMN max_streak INTEGER",
             "ALTER TABLE market_temperature ADD COLUMN broken_rate REAL",
+            "ALTER TABLE recommendations ADD COLUMN entry_signal_type TEXT DEFAULT 'none'",
         ]:
             try:
                 await conn.execute(

backend/app/db/tables.py

@@ -26,7 +26,8 @@ recommendations_table = Table(
     Column("stop_loss", Float),
     Column("reasons", Text),
     Column("llm_analysis", Text, default=""),
-    Column("strategy", Text, default="momentum"),
+    Column("strategy", Text, default="trend_breakout"),
+    Column("entry_signal_type", Text, default="none"),
     Column("llm_score", Float, default=None),
     Column("scan_session", Text),
     Column("created_at", DateTime, server_default=func.now()),

backend/app/engine/recommender.py

@@ -60,17 +60,98 @@ async def get_latest_sectors() -> list[SectorInfo]:
 
 
 async def get_recommendation_history(days: int = 7) -> list[dict]:
-    """获取历史推荐记录"""
+    """获取历史推荐记录，按日期分组返回"""
     from datetime import timedelta
+    import json
+
     start = (datetime.now() - timedelta(days=days)).strftime("%Y-%m-%d")
     async with get_db() as db:
-        from sqlalchemy import select, text
+        from sqlalchemy import text
+
+        # 查询所有历史推荐，按 ts_code 去重（每天取最新一条）
         stmt = text(
-            "SELECT * FROM recommendations WHERE created_at >= :start ORDER BY created_at DESC"
+            "SELECT * FROM recommendations "
+            "WHERE created_at >= :start "
+            "AND id IN ("
+            "  SELECT MAX(id) FROM recommendations "
+            "  WHERE created_at >= :start "
+            "  GROUP BY date(created_at), ts_code"
+            ") "
+            "ORDER BY created_at DESC, score DESC"
         )
         result = await db.execute(stmt, {"start": start})
         rows = result.fetchall()
-        return [dict(row._mapping) for row in rows]
+
+        # 按日期分组
+        grouped: dict[str, list[dict]] = {}
+        for row in rows:
+            r = row._mapping
+            # SQLite created_at 是字符串 "YYYY-MM-DD HH:MM:SS"
+            ca = r["created_at"]
+            if ca:
+                date_str = str(ca)[:10]  # 取前10字符即日期部分
+                created_at_str = str(ca)
+            else:
+                date_str = "unknown"
+                created_at_str = None
+
+            rec_dict = {
+                "ts_code": r["ts_code"],
+                "name": r["name"],
+                "sector": r["sector"] or "",
+                "score": r["score"] or 0,
+                "level": _score_to_level_static(r["score"] or 0),
+                "signal": r["signal"] or "HOLD",
+                "market_temp_score": r["market_temp_score"] or 0,
+                "sector_score": r["sector_score"] or 0,
+                "capital_score": r["capital_score"] or 0,
+                "technical_score": r["technical_score"] or 0,
+                "position_score": r.get("position_score") or 50,
+                "valuation_score": r.get("valuation_score") or 50,
+                "entry_price": r["entry_price"],
+                "target_price": r["target_price"],
+                "stop_loss": r["stop_loss"],
+                "reasons": json.loads(r["reasons"]) if r["reasons"] else [],
+                "risk_note": "",
+                "strategy": r.get("strategy") or "trend_breakout",
+                "entry_signal_type": r.get("entry_signal_type") or "none",
+                "llm_analysis": r.get("llm_analysis") or "",
+                "llm_score": r.get("llm_score"),
+                "scan_session": r["scan_session"] or "",
+                "created_at": created_at_str,
+            }
+
+            if date_str not in grouped:
+                grouped[date_str] = []
+            grouped[date_str].append(rec_dict)
+
+        # 转为列表，按日期降序
+        result_list = []
+        for date_str in sorted(grouped.keys(), reverse=True):
+            recs = grouped[date_str]
+            buy_count = sum(1 for r in recs if r["signal"] == "BUY")
+            avg_score = round(sum(r["score"] for r in recs) / len(recs), 1) if recs else 0
+            result_list.append({
+                "date": date_str,
+                "count": len(recs),
+                "buy_count": buy_count,
+                "avg_score": avg_score,
+                "recommendations": recs,
+            })
+
+        return result_list
+
+
+def _score_to_level_static(score: float) -> str:
+    """根据评分确定推荐等级"""
+    if score >= 75:
+        return "强烈推荐"
+    elif score >= 60:
+        return "推荐"
+    elif score >= 45:
+        return "关注"
+    else:
+        return "观望"
 
 
 async def _save_to_db(result: dict):
@@ -118,7 +199,12 @@ async def _save_to_db(result: dict):
                 )
                 await db.execute(stmt)
 
-            # 保存推荐
+            # 保存推荐（先清除今日旧推荐，避免重复）
+            today_str = datetime.now().strftime("%Y-%m-%d")
+            await db.execute(
+                text("DELETE FROM recommendations WHERE date(created_at) = :today"),
+                {"today": today_str},
+            )
             import json
             for rec in result.get("recommendations", []):
                 stmt = tables.recommendations_table.insert().values(
@@ -139,6 +225,7 @@ async def _save_to_db(result: dict):
                     reasons=json.dumps(rec.reasons, ensure_ascii=False),
                     llm_analysis=rec.llm_analysis,
                     strategy=rec.strategy,
+                    entry_signal_type=rec.entry_signal_type,
                     llm_score=rec.llm_score,
                     scan_session=rec.scan_session,
                 )
@@ -177,9 +264,11 @@ async def _load_today_from_db() -> dict:
                     temperature=m["temperature"],
                 )
 
-            # 加载推荐
+            # 加载推荐（按 ts_code 去重，取最新一条）
             result = await db.execute(
-                text("SELECT * FROM recommendations WHERE date(created_at) = :today ORDER BY score DESC"),
+                text("SELECT * FROM recommendations WHERE date(created_at) = :today "
+                     "AND id IN (SELECT MAX(id) FROM recommendations WHERE date(created_at) = :today GROUP BY ts_code) "
+                     "ORDER BY score DESC"),
                 {"today": today}
             )
             rows = result.fetchall()
@@ -203,7 +292,8 @@ async def _load_today_from_db() -> dict:
                     stop_loss=r["stop_loss"],
                     reasons=json.loads(r["reasons"]) if r["reasons"] else [],
                     llm_analysis=r.get("llm_analysis") or "",
-                    strategy=r.get("strategy") or "momentum",
+                    strategy=r.get("strategy") or "trend_breakout",
+                    entry_signal_type=r.get("entry_signal_type") or "none",
                     llm_score=r.get("llm_score"),
                     scan_session=r["scan_session"] or "",
                 ))

backend/app/engine/screener.py

@@ -1,7 +1,7 @@
-"""双通道漏斗筛选器
+"""趋势突破统一筛选器
 
-Channel A（强中选强）：市场温度 → 板块热度 → 资金筛选 → 技术信号
-Channel B（潜在启动）：全市场技术扫描 → 底部形态 → 估值筛选
+三阶段管道：全市场批量预筛 → 资金流过滤 → 逐股深度分析
+评分公式：趋势&时机30% + 资金流25% + 供需20% + 板块共振15% + 市场温度10%
 
 自动检测是否在交易时段：
   - 盘中模式：用前一日 Tushare 数据 + 腾讯实时行情混合筛选
@@ -12,22 +12,17 @@ import logging
 
 from app.analysis.market_temp import calculate_market_temperature
 from app.analysis.sector_scanner import scan_hot_sectors
-from app.analysis.capital_flow import filter_stocks_by_capital
-from app.analysis.potential_scanner import scan_potential_breakout
+from app.analysis.trend_scanner import scan_trend_breakout
 from app.analysis.signals import generate_signals
-from app.analysis.intraday import intraday_market_temperature, intraday_filter_stocks
+from app.analysis.intraday import intraday_market_temperature, intraday_filter_stocks, intraday_sector_scan
 from app.data.models import MarketTemperature, SectorInfo, TechnicalSignal, Recommendation
-from app.config import settings, is_trading_hours
+from app.config import settings, is_trading_hours, is_market_session
 
 logger = logging.getLogger(__name__)
 
 
 async def run_screening(trade_date: str = None) -> dict:
-    """执行双通道筛选流程
-
-    自动检测交易时段：
-    - 盘中 → 用前一日板块+实时行情筛选
-    - 盘后 → 用当日完整数据筛选
+    """执行趋势突破筛选流程
 
     返回: {
         "market_temp": MarketTemperature,
@@ -36,11 +31,11 @@ async def run_screening(trade_date: str = None) -> dict:
         "scan_mode": "intraday" | "post_market",
     }
     """
-    intraday = is_trading_hours()
+    intraday = is_market_session()
     scan_mode = "intraday" if intraday else "post_market"
     logger.info(f"=== 筛选模式: {'盘中实时' if intraday else '盘后'} ===")
 
-    # ── 市场温度（共享） ──
+    # ── 市场温度 ──
     logger.info("=== 市场温度计 ===")
     market_temp = calculate_market_temperature(trade_date)
 
@@ -52,151 +47,147 @@ async def run_screening(trade_date: str = None) -> dict:
 
     market_temp_score = market_temp.temperature
 
-    # ── 板块热度（Channel A 需要） ──
+    # ── 板块热度（用于板块共振评分） ──
     logger.info("=== 板块热度扫描 ===")
     all_sectors = scan_hot_sectors(trade_date)
     hot_sectors = all_sectors[:settings.top_sector_count]
 
-    # ── Channel A：强中选强 ──
-    recommendations_a = []
-    capital_filtered = []
+    # 盘中用实时行情更新板块涨幅和涨停数
+    if intraday:
+        hot_sectors = await intraday_sector_scan(hot_sectors)
 
-    if hot_sectors:
-        if intraday:
-            logger.info("=== Channel A：盘中实时个股筛选 ===")
-            capital_filtered = await intraday_filter_stocks(hot_sectors)
-        else:
-            logger.info("=== Channel A：个股资金筛选 ===")
-            capital_filtered = await filter_stocks_by_capital(hot_sectors, trade_date)
+    # ── 趋势突破三阶段管道 ──
+    logger.info("=== 趋势突破扫描 ===")
+    candidates = await scan_trend_breakout(
+        trade_date=trade_date,
+        market_temp=market_temp,
+        hot_sectors=hot_sectors,
+        intraday=intraday,
+    )
 
-        if capital_filtered:
-            recommendations_a = _build_recommendations(
-                capital_filtered, market_temp, hot_sectors,
-                market_temp_score=market_temp_score,
-                strategy="momentum", intraday=intraday,
-            )
+    if not candidates:
+        logger.info("=== 筛选完成: 0 只股票 ===")
+        return {
+            "market_temp": market_temp,
+            "hot_sectors": hot_sectors,
+            "recommendations": [],
+            "scan_mode": scan_mode,
+        }
 
-    logger.info(f"Channel A（强中选强）: {len(recommendations_a)} 只")
+    # ── 构建推荐列表 ──
+    recommendations = _build_trend_recommendations(
+        candidates, market_temp, hot_sectors, market_temp_score, intraday,
+    )
 
-    # ── Channel B：潜在启动 ──
-    logger.info("=== Channel B：潜在启动扫描 ===")
-    exclude_codes = {r.ts_code for r in recommendations_a}
-    potential_filtered = scan_potential_breakout(trade_date, exclude_codes)
+    # 过滤低质量推荐
+    recommendations = [r for r in recommendations if r.score >= 40]
 
-    recommendations_b = []
-    if potential_filtered:
-        recommendations_b = _build_recommendations(
-            potential_filtered, market_temp, hot_sectors,
-            market_temp_score=market_temp_score,
-            strategy="potential", intraday=intraday,
-        )
-
-    logger.info(f"Channel B（潜在启动）: {len(recommendations_b)} 只")
-
-    # 合并，按评分排序
-    all_recommendations = recommendations_a + recommendations_b
-    all_recommendations.sort(key=lambda x: x.score, reverse=True)
-
-    # 过滤掉低质量推荐
-    all_recommendations = [r for r in all_recommendations if r.score >= 40]
-
-    logger.info(f"=== 筛选完成: {len(all_recommendations)} 只股票 ({scan_mode}) ===")
-    for r in all_recommendations[:5]:
-        strategy_label = "强中选强" if r.strategy == "momentum" else "潜在启动"
-        logger.info(f"  [{strategy_label}] {r.name}({r.ts_code}) {r.level} 评分={r.score} 信号={r.signal}")
+    logger.info(f"=== 筛选完成: {len(recommendations)} 只股票 ({scan_mode}) ===")
+    for r in recommendations[:5]:
+        signal_map = {"breakout": "突破型", "pullback": "回踩型", "launch": "启动型"}
+        signal_label = signal_map.get(r.entry_signal_type, r.entry_signal_type)
+        logger.info(f"  [{signal_label}] {r.name}({r.ts_code}) {r.level} 评分={r.score} 信号={r.signal}")
 
     return {
         "market_temp": market_temp,
         "hot_sectors": hot_sectors,
-        "capital_filtered": capital_filtered,
-        "recommendations": all_recommendations,
+        "recommendations": recommendations,
         "scan_mode": scan_mode,
     }
 
 
-def _build_recommendations(
-    stocks: list[dict],
+def _build_trend_recommendations(
+    candidates: list[dict],
     market_temp: MarketTemperature,
     hot_sectors: list[SectorInfo],
     market_temp_score: float = 0,
-    strategy: str = "momentum",
     intraday: bool = False,
 ) -> list[Recommendation]:
-    """从筛选结果构建推荐列表（Channel A/B 共用）"""
+    """从趋势突破扫描结果构建推荐列表
+
+    评分公式：趋势&时机30% + 资金流25% + 供需20% + 板块共振15% + 市场温度10%
+    """
     recommendations = []
 
-    for stock in stocks:
+    for stock in candidates:
         ts_code = stock["ts_code"]
         name = stock["name"]
         sector = stock["sector"]
+        entry_signal_type = stock.get("entry_signal_type", "none")
+        entry_signal_score = stock.get("entry_signal_score", 0)
+        tech_signal = stock.get("tech_signal")
 
-        tech_signal = generate_signals(ts_code, name)
-
-        # 板块得分
-        if strategy == "momentum":
-            sector_score = _get_sector_score(sector, hot_sectors)
-            sector_stage = _get_sector_stage(sector, hot_sectors)
-        else:
-            # Channel B：不在热门板块中，给基础分
-            sector_score = 30.0
-            sector_stage = "mid"
-
-        # 估值安全得分
+        # 各维度得分
+        trend_timing_score = stock.get("trend_timing_score", 50)
+        supply_demand_score = stock.get("supply_demand_score", 50)
+        capital_score = stock.get("capital_score", 50)
+        position_score = stock.get("position_score", 50)
         valuation_score = stock.get("valuation_score", 50)
 
-        # 位置安全得分
-        position_score = tech_signal.position_score
+        # 板块共振评分
+        sector_score = _score_sector_resonance(sector, hot_sectors)
+        sector_stage = _get_sector_stage(sector, hot_sectors)
 
-        # 综合评分（根据策略调整权重）
-        if strategy == "momentum":
-            # 强中选强：板块和资金权重高
-            # 市场10% + 板块20% + 资金20% + 技术15% + 位置安全15% + 估值安全20%
-            final_score = (
-                market_temp_score * 0.10 +
-                sector_score * 0.20 +
-                stock["capital_score"] * 0.20 +
-                tech_signal.score * 0.15 +
-                position_score * 0.15 +
-                valuation_score * 0.20
-            )
-        else:
-            # 潜在启动：技术面和估值权重高
-            # 市场10% + 技术25% + 资金(potential_score)15% + 位置安全25% + 估值安全25%
-            final_score = (
-                market_temp_score * 0.10 +
-                tech_signal.score * 0.25 +
-                stock["capital_score"] * 0.15 +
-                position_score * 0.25 +
-                valuation_score * 0.25
-            )
+        # 综合评分（新权重）
+        final_score = (
+            trend_timing_score * 0.30 +
+            capital_score * 0.25 +
+            supply_demand_score * 0.20 +
+            sector_score * 0.15 +
+            market_temp_score * 0.10
+        )
 
-        # 板块尾声阶段额外惩罚（仅 Channel A）
-        if strategy == "momentum":
-            if sector_stage == "end":
-                final_score *= 0.85
-            elif sector_stage == "late":
-                final_score *= 0.92
+        # 风险乘数
+        if tech_signal:
+            if tech_signal.rally_pct_5d > 20:
+                final_score *= 0.65
+            elif tech_signal.rally_pct_5d > 15:
+                final_score *= 0.80
+
+        if sector_stage == "end":
+            final_score *= 0.70
+        elif sector_stage == "late":
+            final_score *= 0.88
+
+        if market_temp_score < 30:
+            final_score *= 0.75
+
+        # 入场信号高置信度奖励
+        if entry_signal_score >= 80:
+            final_score *= 1.10
 
         # 确定信号和等级
-        signal = "HOLD"
-        if strategy == "momentum":
-            if (tech_signal.score >= settings.buy_score_threshold
-                    and tech_signal.signal_count >= settings.buy_min_signals
-                    and position_score >= 30):
-                signal = "BUY"
-        else:
-            # Channel B：技术面要求稍低，但位置安全要求更高
-            if (tech_signal.score >= settings.buy_score_threshold * 0.7
-                    and position_score >= 40):
-                signal = "BUY"
-
         level = _score_to_level(final_score)
+        signal = "HOLD"
+        if entry_signal_type != "none" and entry_signal_score >= 50 and position_score >= 30 and final_score >= 60:
+            signal = "BUY"
+
+        # 价格参考
+        entry_price = None
+        target_price = None
+        stop_loss = None
+        if tech_signal:
+            entry_price = tech_signal.support_price
+            target_price = tech_signal.resist_price
+            stop_loss = tech_signal.stop_loss_price
+
+            # 根据入场信号类型调整参考价
+            details = stock.get("entry_signal_details", {})
+            if entry_signal_type == "breakout" and details.get("resist_level"):
+                entry_price = details["resist_level"]
+                target_price = round(entry_price * 1.05, 2)
+            elif entry_signal_type == "pullback" and details.get("support_price"):
+                entry_price = details["support_price"]
+                target_price = round(entry_price * 1.05, 2)
+            elif entry_signal_type == "launch" and details.get("resist_level"):
+                entry_price = round(details["resist_level"] * 1.01, 2)
+                target_price = round(details["resist_level"] * 1.08, 2)
 
         # 生成推荐理由
-        reasons = _generate_reasons(stock, tech_signal, market_temp, intraday, strategy)
+        reasons = _generate_reasons(stock, tech_signal, market_temp, intraday)
 
         # 风险提示
-        risk_note = _generate_risk_note(market_temp, tech_signal, intraday, strategy)
+        risk_note = _generate_risk_note(market_temp, tech_signal, stock)
 
         rec = Recommendation(
             ts_code=ts_code,
@@ -205,24 +196,41 @@ def _build_recommendations(
             score=round(final_score, 1),
             market_temp_score=round(market_temp_score, 1),
             sector_score=round(sector_score, 1),
-            capital_score=round(stock["capital_score"], 1),
-            technical_score=round(tech_signal.score, 1),
+            capital_score=round(capital_score, 1),
+            technical_score=round(stock.get("technical_score", 50), 1),
             position_score=round(position_score, 1),
             valuation_score=round(valuation_score, 1),
             signal=signal,
-            entry_price=tech_signal.support_price,
-            target_price=tech_signal.resist_price,
-            stop_loss=tech_signal.stop_loss_price,
+            entry_price=entry_price,
+            target_price=target_price,
+            stop_loss=stop_loss,
             reasons=reasons,
             risk_note=risk_note,
             level=level,
-            strategy=strategy,
+            strategy="trend_breakout",
+            entry_signal_type=entry_signal_type,
         )
         recommendations.append(rec)
 
     return recommendations
 
 
+def _score_sector_resonance(sector_name: str, hot_sectors: list[SectorInfo]) -> float:
+    """板块共振评分 (0-100)"""
+    for s in hot_sectors:
+        if s.sector_name == sector_name:
+            score = 40  # 在热门板块列表中
+            score += s.heat_score * 0.3  # 板块热度贡献
+            if s.stage == "early":
+                score += 30
+            elif s.stage == "mid":
+                score += 20
+            elif s.stage == "late":
+                score += 5
+            return min(score, 100)
+    return 10.0  # 不在热门板块
+
+
 def _get_sector_score(sector_name: str, hot_sectors: list[SectorInfo]) -> float:
     """获取板块在热门板块中的得分"""
     for s in hot_sectors:
@@ -251,99 +259,93 @@ def _score_to_level(score: float) -> str:
 
 
 def _generate_reasons(
-    stock: dict, tech: TechnicalSignal, market: MarketTemperature,
-    intraday: bool = False, strategy: str = "momentum",
+    stock: dict, tech: TechnicalSignal | None,
+    market: MarketTemperature, intraday: bool = False,
 ) -> list[str]:
+    """生成推荐理由"""
     reasons = []
+    entry_type = stock.get("entry_signal_type", "none")
+    signal_map = {"breakout": "突破型", "pullback": "回踩型", "launch": "启动型"}
+    entry_label = signal_map.get(entry_type, "")
 
-    if strategy == "potential":
-        # Channel B 理由：侧重底部形态和估值
-        if tech.position_score >= 70:
-            reasons.append("位置处于低位，距高点回调充分")
-        if tech.macd_golden:
-            reasons.append("MACD底部金叉，反转信号初现")
-        if tech.pullback_support:
-            reasons.append("缩量回踩支撑位，蓄势充分")
-        if tech.big_yang:
-            reasons.append("底部出现放量长阳，资金介入")
-        if stock.get("valuation_score", 0) >= 60:
-            reasons.append("估值安全，下行空间有限")
-        if not reasons:
-            reasons.append("技术面底部信号显现，关注启动时机")
-    elif intraday:
-        # Channel A 盘中理由
-        pct = stock.get("pct_chg", 0)
-        vr = stock.get("volume_ratio")
-        if vr and vr > 2:
-            reasons.append(f"盘中量比{vr:.1f}倍，资金活跃度高")
-        if pct > 3:
-            reasons.append(f"盘中涨幅{pct:.1f}%，走势强劲")
-        elif pct > 0:
-            reasons.append(f"盘中涨幅{pct:.1f}%，温和上攻")
+    # 入场信号
+    if entry_label:
+        details = stock.get("entry_signal_details", {})
+        if entry_type == "breakout":
+            breakout_pct = details.get("breakout_pct", 0)
+            vol_ratio = details.get("volume_ratio", 0)
+            reasons.append(f"放量突破20日阻力位（涨幅{breakout_pct:.1f}%，量比{vol_ratio:.1f}倍）")
+        elif entry_type == "pullback":
+            support = details.get("support_ma", "")
+            shrink = details.get("volume_shrink_ratio", 0)
+            reasons.append(f"缩量回踩{support}支撑（量能收缩至{shrink:.0%}）")
+        elif entry_type == "launch":
+            range_pct = details.get("price_range_pct", 0)
+            shrink = details.get("volume_shrink_ratio", 0)
+            reasons.append(f"高位缩量整理{range_pct:.1f}%后即将变盘（量缩至{shrink:.0%}）")
 
-        reasons.append(f"所属板块【{stock['sector']}】为当前热门概念")
+    # 供需分析
+    vol_trend = stock.get("volume_trend", "")
+    ds_ratio = stock.get("demand_supply_ratio", 1)
+    if ds_ratio > 1.5:
+        reasons.append(f"需求主导（上涨均量/下跌均量={ds_ratio:.1f}）")
+    elif vol_trend == "expanding":
+        reasons.append("量能逐步放大，资金持续介入")
 
+    # 资金流
+    main_net = stock.get("main_net_inflow", 0)
+    if main_net > 5000:
+        reasons.append(f"主力资金大幅流入{main_net:.0f}万元")
+    elif main_net > 1000:
+        reasons.append(f"主力资金持续流入{main_net:.0f}万元")
+
+    # 板块
+    sector = stock.get("sector", "")
+    if sector:
+        reasons.append(f"所属板块【{sector}】")
+
+    # 技术面
+    if tech:
         tech_reasons = []
         if tech.ma_bullish:
             tech_reasons.append("均线多头排列")
-        if tech.volume_breakout:
-            tech_reasons.append("放量突破")
-        if tech.macd_golden:
-            tech_reasons.append("MACD金叉")
-        if tech_reasons:
-            reasons.append("技术面: " + "、".join(tech_reasons))
-    else:
-        # Channel A 盘后理由
-        inflow = stock.get("main_net_inflow", 0)
-        if inflow > 5000:
-            reasons.append(f"主力资金大幅流入{inflow:.0f}万元")
-        elif inflow > 1000:
-            reasons.append(f"主力资金持续流入{inflow:.0f}万元")
-
-        reasons.append(f"所属板块【{stock['sector']}】为当前热门概念")
-
-        tech_reasons = []
-        if tech.ma_bullish:
-            tech_reasons.append("均线多头排列")
-        if tech.volume_breakout:
-            tech_reasons.append("放量突破")
         if tech.macd_golden:
             tech_reasons.append("MACD金叉")
         if tech.pullback_support:
             tech_reasons.append("缩量回踩支撑")
-        if tech.big_yang:
-            tech_reasons.append("底部放量长阳")
         if tech_reasons:
             reasons.append("技术面: " + "、".join(tech_reasons))
 
-    # 位置安全
-    if tech.position_score >= 70:
-        reasons.append("位置安全，距高点有空间")
-    elif tech.position_score < 30:
-        reasons.append("注意：短期涨幅较大，追高风险")
-
     return reasons[:3]
 
 
 def _generate_risk_note(
-    market: MarketTemperature, tech: TechnicalSignal,
-    intraday: bool = False, strategy: str = "momentum",
+    market: MarketTemperature,
+    tech: TechnicalSignal | None,
+    stock: dict,
 ) -> str:
+    """生成风险提示"""
     notes = []
-    if intraday:
-        notes.append("盘中数据参考，需结合尾盘确认")
-    if strategy == "potential":
-        notes.append("底部股票可能继续盘整，注意时间成本")
+    entry_type = stock.get("entry_signal_type", "")
+
+    if entry_type == "breakout":
+        notes.append("突破型需警惕假突破，关注量能是否持续")
+    elif entry_type == "pullback":
+        notes.append("回踩型可能继续下探支撑，注意止损纪律")
+    elif entry_type == "launch":
+        notes.append("启动型整理可能延长，注意时间成本")
+
     if market.temperature < 30:
         notes.append("市场情绪偏冷，系统性风险较高")
     elif market.temperature < 50:
         notes.append("市场情绪一般，注意仓位控制")
-    if tech.score < 40:
-        notes.append("技术面支撑不足")
-    if tech.position_score < 30:
-        notes.append(f"近期涨幅较大(5日{tech.rally_pct_5d}%)，追高风险")
-    if tech.rally_pct_10d > 20:
-        notes.append(f"10日累涨{tech.rally_pct_10d}%，警惕回调")
+
+    if tech:
+        if tech.position_score < 30:
+            notes.append(f"近期涨幅较大(5日{tech.rally_pct_5d}%)，追高风险")
+        if tech.rally_pct_10d > 20:
+            notes.append(f"10日累涨{tech.rally_pct_10d}%，警惕回调")
+
     if not notes:
         return "注意设好止损，控制仓位"
     return "；".join(notes)

backend/app/llm/analysis_agent.py

@@ -10,10 +10,7 @@ import logging
 import re
 
 from app.llm.client import chat_completion
-from app.llm.prompts import (
-    MOMENTUM_ANALYSIS_PROMPT,
-    POTENTIAL_ANALYSIS_PROMPT,
-)
+from app.llm.prompts import TREND_BREAKOUT_ANALYSIS_PROMPT
 from app.config import settings
 
 logger = logging.getLogger(__name__)
@@ -62,16 +59,15 @@ async def _do_analyze(result: dict, recommendations: list) -> None:
             # 预先获取该股票的详细数据
             stock_data = await _fetch_stock_data(rec.ts_code, rec.sector)
 
-            strategy_label = "强中选强" if rec.strategy == "momentum" else "潜在启动"
-            system_prompt = (
-                MOMENTUM_ANALYSIS_PROMPT
-                if rec.strategy == "momentum"
-                else POTENTIAL_ANALYSIS_PROMPT
-            )
+            strategy_label = "趋势突破"
+            signal_map = {"breakout": "突破型", "pullback": "回踩型", "launch": "启动型", "none": "无信号"}
+            entry_label = signal_map.get(rec.entry_signal_type, "无信号")
+            system_prompt = TREND_BREAKOUT_ANALYSIS_PROMPT
 
             user_msg = _build_user_message(
                 rec=rec,
                 strategy_label=strategy_label,
+                entry_label=entry_label,
                 market_temp=market_temp,
                 temp_level=temp_level,
                 sectors_text=sectors_text,
@@ -169,6 +165,7 @@ async def _fetch_stock_data(ts_code: str, sector: str) -> str:
 def _build_user_message(
     rec,
     strategy_label: str,
+    entry_label: str,
     market_temp,
     temp_level: str,
     sectors_text: str,
@@ -179,6 +176,7 @@ def _build_user_message(
 - 股票: {rec.name}({rec.ts_code})
 - 所属板块: {rec.sector}
 - 策略类型: {strategy_label}
+- 入场信号: {entry_label}
 - 综合评分: {rec.score}分（{rec.level}）
 - 各维度: 市场{rec.market_temp_score} | 板块{rec.sector_score} | 资金{rec.capital_score} | 技术{rec.technical_score} | 位置{rec.position_score} | 估值{rec.valuation_score}
 - 信号: {rec.signal}

backend/app/llm/prompts.py

@@ -54,63 +54,42 @@ CHAT_SYSTEM_PROMPT = """\
 
 # ── AI 分析 Agent Prompt ──
 
-MOMENTUM_ANALYSIS_PROMPT = """\
-你是一位专业的 A 股趋势交易分析师。你需要评估一只处于热门板块中的强势股是否值得追入。
+TREND_BREAKOUT_ANALYSIS_PROMPT = """\
+你是一位专业的 A 股趋势突破交易分析师。你需要评估一只处于上升趋势中、即将突破的股票的入场时机。
 
 系统已为你提供了该股票的量化评分、K线数据、资金流向、技术信号、板块数据等详细信息，请基于这些数据进行深度分析。
 
 重点关注：
-1. 当前趋势的持续性：量价是否配合？资金流入是否持续？
-2. 追入的安全性：当前位置高低？短期是否过热？
-3. 入场时机：应该回调到支撑位买入，还是突破追入？
-4. 风险收益比：上行目标空间 vs 下行止损空间
+1. 入场信号类型确认：突破型/回踩型/启动型，信号是否可靠？
+2. 量价配合：上涨放量、回调缩量的特征是否明显？
+3. 资金持续性：主力资金是否持续流入3天以上？
+4. 1-5日操作策略：最佳入场价位、目标价位（2-5%空间）、止损价位
+5. 时机判断：预计突破/反弹在1-3天内发生的概率
 
 请严格按以下格式输出分析报告：
 
-### 核心逻辑
-（1-2句核心投资逻辑，说明为什么值得关注）
+### 信号类型确认
+（突破型/回踩型/启动型，判断依据，可靠性评估）
 
-### 趋势分析
-（均线排列、MACD状态、成交量变化趋势，用数据说话）
+### 量价分析
+（成交量变化趋势、供需关系、资金介入程度）
 
-### 入场策略
-（建议的入场价位和方式：回调买入/突破买入/分批建仓）
+### 操作策略（1-5日）
+（入场价位、分批建仓计划、目标价位、止损价位）
+
+### 时间窗口
+（预计启动时间、关键观察节点）
 
 ### 风险提示
-（主要风险因素：板块衰退、大盘系统性风险、量能不济等）
+（主要风险因素：假突破风险、板块衰退、大盘系统性风险等）
 
 ### AI 评分
-（给出 1-10 分，格式为纯数字，如：7）
+（给出 1-10 分，格式为纯数字，如：8）
 """
 
-POTENTIAL_ANALYSIS_PROMPT = """\
-你是一位专业的 A 股底部反转交易分析师。你需要评估一只处于底部的股票是否具备启动条件。
-
-系统已为你提供了该股票的量化评分、K线数据、资金流向、技术信号、板块数据等详细信息，请基于这些数据进行深度分析。
-
-重点关注：
-1. 底部信号的可信度：量价配合如何？多个技术指标是否共振？
-2. 可能的催化剂：板块轮动机会？资金是否有流入迹象？技术面是否接近突破？
-3. 时间窗口：底部蓄势了多久？均线是否收敛？何时可能启动？
-4. 风险：继续下行的概率和幅度？是否有基本面隐患？
-
-请严格按以下格式输出分析报告：
-
-### 底部信号分析
-（底部形态特征、量价变化、技术指标状态）
-
-### 催化剂判断
-（可能的上涨催化剂：板块轮动、资金流入、技术突破等）
-
-### 埋伏策略
-（建议的建仓方式和价位、仓位建议、等待确认信号）
-
-### 风险提示
-（主要风险因素：继续下行风险、底部无效风险、时间成本等）
-
-### AI 评分
-（给出 1-10 分，格式为纯数字，如：6）
-"""
+# 保留旧 prompt 用于向后兼容（旧推荐数据仍可能使用）
+MOMENTUM_ANALYSIS_PROMPT = TREND_BREAKOUT_ANALYSIS_PROMPT
+POTENTIAL_ANALYSIS_PROMPT = TREND_BREAKOUT_ANALYSIS_PROMPT
 
 ANALYSIS_USER_TEMPLATE = """\
 ## 量化系统数据

frontend/.next/app-build-manifest.json

@@ -1,5 +1,10 @@
 {
   "pages": {
+    "/sectors/page": [
+      "static/chunks/webpack.js",
+      "static/chunks/main-app.js",
+      "static/chunks/app/sectors/page.js"
+    ],
     "/layout": [
       "static/chunks/webpack.js",
       "static/chunks/main-app.js",
@@ -11,10 +16,10 @@
       "static/chunks/main-app.js",
       "static/chunks/app/page.js"
     ],
-    "/stock/[code]/page": [
+    "/recommendations/page": [
       "static/chunks/webpack.js",
       "static/chunks/main-app.js",
-      "static/chunks/app/stock/[code]/page.js"
+      "static/chunks/app/recommendations/page.js"
     ]
   }
 }

frontend/.next/build-manifest.json

@@ -2,9 +2,7 @@
   "polyfillFiles": [
     "static/chunks/polyfills.js"
   ],
-  "devFiles": [
-    "static/chunks/react-refresh.js"
-  ],
+  "devFiles": [],
   "ampDevFiles": [],
   "lowPriorityFiles": [
     "static/development/_buildManifest.js",
@@ -15,16 +13,7 @@
     "static/chunks/main-app.js"
   ],
   "pages": {
-    "/_app": [
-      "static/chunks/webpack.js",
-      "static/chunks/main.js",
-      "static/chunks/pages/_app.js"
-    ],
-    "/_error": [
-      "static/chunks/webpack.js",
-      "static/chunks/main.js",
-      "static/chunks/pages/_error.js"
-    ]
+    "/_app": []
   },
   "ampFirstPages": []
 }

frontend/.next/react-loadable-manifest.json

@@ -1,20 +1 @@
-{
-  "components/capital-flow.tsx -> echarts": {
-    "id": "components/capital-flow.tsx -> echarts",
-    "files": [
-      "static/chunks/_app-pages-browser_node_modules_echarts_index_js.js"
-    ]
-  },
-  "components/kline-chart.tsx -> echarts": {
-    "id": "components/kline-chart.tsx -> echarts",
-    "files": [
-      "static/chunks/_app-pages-browser_node_modules_echarts_index_js.js"
-    ]
-  },
-  "components/score-radar.tsx -> echarts": {
-    "id": "components/score-radar.tsx -> echarts",
-    "files": [
-      "static/chunks/_app-pages-browser_node_modules_echarts_index_js.js"
-    ]
-  }
-}
+{}

frontend/.next/server/app-paths-manifest.json

@@ -1,4 +1,5 @@
 {
-  "/stock/[code]/page": "app/stock/[code]/page.js",
-  "/page": "app/page.js"
+  "/recommendations/page": "app/recommendations/page.js",
+  "/page": "app/page.js",
+  "/sectors/page": "app/sectors/page.js"
 }

frontend/.next/server/middleware-build-manifest.js

@@ -2,9 +2,7 @@ self.__BUILD_MANIFEST = {
   "polyfillFiles": [
     "static/chunks/polyfills.js"
   ],
-  "devFiles": [
-    "static/chunks/react-refresh.js"
-  ],
+  "devFiles": [],
   "ampDevFiles": [],
   "lowPriorityFiles": [],
   "rootMainFiles": [
@@ -12,16 +10,7 @@ self.__BUILD_MANIFEST = {
     "static/chunks/main-app.js"
   ],
   "pages": {
-    "/_app": [
-      "static/chunks/webpack.js",
-      "static/chunks/main.js",
-      "static/chunks/pages/_app.js"
-    ],
-    "/_error": [
-      "static/chunks/webpack.js",
-      "static/chunks/main.js",
-      "static/chunks/pages/_error.js"
-    ]
+    "/_app": []
   },
   "ampFirstPages": []
 };

frontend/.next/server/middleware-react-loadable-manifest.js

@@ -1 +1 @@
-self.__REACT_LOADABLE_MANIFEST="{\"components/capital-flow.tsx -> echarts\":{\"id\":\"components/capital-flow.tsx -> echarts\",\"files\":[\"static/chunks/_app-pages-browser_node_modules_echarts_index_js.js\"]},\"components/kline-chart.tsx -> echarts\":{\"id\":\"components/kline-chart.tsx -> echarts\",\"files\":[\"static/chunks/_app-pages-browser_node_modules_echarts_index_js.js\"]},\"components/score-radar.tsx -> echarts\":{\"id\":\"components/score-radar.tsx -> echarts\",\"files\":[\"static/chunks/_app-pages-browser_node_modules_echarts_index_js.js\"]}}"
+self.__REACT_LOADABLE_MANIFEST="{}"

frontend/.next/server/pages-manifest.json

@@ -1,5 +1 @@
-{
-  "/_app": "pages/_app.js",
-  "/_error": "pages/_error.js",
-  "/_document": "pages/_document.js"
-}
+{}

frontend/.next/server/server-reference-manifest.json

@@ -1,5 +1,5 @@
 {
   "node": {},
   "edge": {},
-  "encryptionKey": "b90zeeYzMcSqNivimWn6T2PQQNxojiQrSeFlOxUaP4Q="
+  "encryptionKey": "IAps6Nn+QS9GVH+sOr6laVRGfUJrD0aLxGy9A/+XkIs="
 }

frontend/.next/server/webpack-runtime.js

@@ -125,7 +125,7 @@
 /******/ 	
 /******/ 	/* webpack/runtime/getFullHash */
 /******/ 	(() => {
-/******/ 		__webpack_require__.h = () => ("64fc734a2bef9f33")
+/******/ 		__webpack_require__.h = () => ("a035f49818643978")
 /******/ 	})();
 /******/ 	
 /******/ 	/* webpack/runtime/hasOwnProperty shorthand */

frontend/src/app/page.tsx

@@ -7,6 +7,7 @@ import MarketTemp from "@/components/market-temp";
 import StockCard from "@/components/stock-card";
 import SectorHeatmap from "@/components/sector-heatmap";
 import { useWebSocket } from "@/hooks/use-websocket";
+import { useAuth } from "@/hooks/use-auth";
 
 interface ScanStatus {
   is_trading: boolean;
@@ -15,6 +16,7 @@ interface ScanStatus {
 }
 
 export default function DashboardPage() {
+  const { user } = useAuth();
   const [data, setData] = useState<LatestResult | null>(null);
   const [sectors, setSectors] = useState<SectorData[]>([]);
   const [scanStatus, setScanStatus] = useState<ScanStatus | null>(null);
@@ -114,6 +116,7 @@ export default function DashboardPage() {
             </p>
           )}
         </div>
+        {user?.role === "admin" && (
         <button
           onClick={handleRefresh}
           disabled={refreshing}
@@ -130,6 +133,7 @@ export default function DashboardPage() {
             "立即扫描"
           )}
         </button>
+        )}
       </div>
 
       {/* Scan result toast */}
@@ -167,7 +171,7 @@ export default function DashboardPage() {
           <div className="glass-card-static p-10 text-center">
             <div className="text-text-muted text-sm mb-1">暂无推荐</div>
             <div className="text-text-muted/60 text-xs">
-              点击 {scanStatus?.is_trading ? "「盘中扫描」" : "「立即扫描」"} 开始分析
+              {user?.role === "admin" ? `点击 ${scanStatus?.is_trading ? "「盘中扫描」" : "「立即扫描」"} 开始分析` : "等待系统自动扫描更新"}
             </div>
           </div>
         ) : (

frontend/src/app/recommendations/page.tsx

@@ -2,12 +2,26 @@
 
 import { useEffect, useState, useCallback } from "react";
 import { fetchAPI, postAPI } from "@/lib/api";
-import type { LatestResult } from "@/lib/api";
+import type { DayGroup } from "@/lib/api";
 import StockCard from "@/components/stock-card";
 import { useWebSocket } from "@/hooks/use-websocket";
 
+function formatDate(dateStr: string): string {
+  const d = new Date(dateStr);
+  const today = new Date();
+  const yesterday = new Date(today);
+  yesterday.setDate(yesterday.getDate() - 1);
+
+  if (dateStr === today.toISOString().slice(0, 10)) return "今日";
+  if (dateStr === yesterday.toISOString().slice(0, 10)) return "昨日";
+
+  const weekDays = ["周日", "周一", "周二", "周三", "周四", "周五", "周六"];
+  return `${d.getMonth() + 1}月${d.getDate()}日 ${weekDays[d.getDay()]}`;
+}
+
 export default function RecommendationsPage() {
-  const [data, setData] = useState<LatestResult | null>(null);
+  const [dayGroups, setDayGroups] = useState<DayGroup[]>([]);
+  const [expandedDays, setExpandedDays] = useState<Set<string>>(new Set());
   const [filter, setFilter] = useState<string>("all");
   const [llmEnabled, setLlmEnabled] = useState(false);
   const [refreshing, setRefreshing] = useState(false);
@@ -15,12 +29,22 @@ export default function RecommendationsPage() {
 
   const loadData = useCallback(async () => {
     try {
-      const [result, health] = await Promise.all([
-        fetchAPI<LatestResult>("/api/recommendations/latest"),
+      const [history, health] = await Promise.all([
+        fetchAPI<DayGroup[]>("/api/recommendations/history?days=14"),
         fetchAPI<{ llm_enabled: boolean }>("/api/health"),
       ]);
-      setData(result);
+      setDayGroups(history);
       setLlmEnabled(health.llm_enabled);
+
+      // 默认展开今天
+      const today = new Date().toISOString().slice(0, 10);
+      setExpandedDays((prev) => {
+        const next = new Set(prev);
+        if (!next.size && history.length > 0) {
+          next.add(history[0].date);
+        }
+        return next;
+      });
     } catch (e) {
       console.error("加载推荐失败:", e);
     }
@@ -59,17 +83,26 @@ export default function RecommendationsPage() {
     }
   };
 
-  const recs = data?.recommendations ?? [];
-  const filtered =
-    filter === "all"
-      ? recs
-      : filter === "buy"
-      ? recs.filter((r) => r.signal === "BUY")
-      : filter === "momentum"
-      ? recs.filter((r) => r.strategy === "momentum")
-      : filter === "potential"
-      ? recs.filter((r) => r.strategy === "potential")
-      : recs.filter((r) => r.level === filter);
+  const toggleDay = (date: string) => {
+    setExpandedDays((prev) => {
+      const next = new Set(prev);
+      if (next.has(date)) next.delete(date);
+      else next.add(date);
+      return next;
+    });
+  };
+
+  const applyFilter = (recs: DayGroup["recommendations"]) => {
+    if (filter === "all") return recs;
+    if (filter === "buy") return recs.filter((r) => r.signal === "BUY");
+    if (filter === "breakout") return recs.filter((r) => r.entry_signal_type === "breakout");
+    if (filter === "pullback") return recs.filter((r) => r.entry_signal_type === "pullback");
+    if (filter === "launch") return recs.filter((r) => r.entry_signal_type === "launch");
+    return recs.filter((r) => r.level === filter);
+  };
+
+  // 总数统计
+  const totalCount = dayGroups.reduce((sum, g) => sum + applyFilter(g.recommendations).length, 0);
 
   return (
     <div className="max-w-7xl mx-auto px-4 md:px-8 pt-6 pb-20 md:pb-10">
@@ -78,7 +111,8 @@ export default function RecommendationsPage() {
         <div>
           <h1 className="text-lg font-bold tracking-tight">推荐列表</h1>
           <p className="text-xs text-text-muted mt-0.5">
-            共 <span className="font-mono tabular-nums">{filtered.length}</span> 只
+            共 <span className="font-mono tabular-nums">{totalCount}</span> 只 ·
+            <span className="font-mono tabular-nums ml-1">{dayGroups.length}</span> 天记录
           </p>
         </div>
         <button
@@ -109,12 +143,12 @@ export default function RecommendationsPage() {
       <div className="flex gap-2 mb-5 overflow-x-auto pb-1 animate-fade-in-up delay-75">
         {[
           { key: "all", label: "全部" },
-          { key: "momentum", label: "强中选强" },
-          { key: "potential", label: "潜在启动" },
+          { key: "breakout", label: "突破型" },
+          { key: "pullback", label: "回踩型" },
+          { key: "launch", label: "启动型" },
           { key: "buy", label: "买入信号" },
           { key: "强烈推荐", label: "强烈推荐" },
           { key: "推荐", label: "推荐" },
-          { key: "观望", label: "观望" },
         ].map(({ key, label }) => (
           <button
             key={key}
@@ -130,18 +164,96 @@ export default function RecommendationsPage() {
         ))}
       </div>
 
-      {filtered.length === 0 ? (
+      {/* Day groups */}
+      {dayGroups.length === 0 ? (
         <div className="glass-card-static p-12 text-center animate-fade-in-up">
-          <div className="text-text-muted text-sm mb-1">暂无{filter === "all" ? "" : "符合条件的"}推荐</div>
-          <div className="text-text-muted/50 text-xs">尝试切换筛选条件或触发新的扫描</div>
+          <div className="text-text-muted text-sm mb-1">暂无推荐记录</div>
+          <div className="text-text-muted/50 text-xs">触发新的扫描以生成推荐</div>
         </div>
       ) : (
-        <div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-4">
-          {filtered.map((rec, i) => (
-            <div key={rec.ts_code} className="animate-fade-in-up" style={{ animationDelay: `${i * 60}ms` }}>
-              <StockCard rec={rec} showLLMLoading={llmEnabled} />
-            </div>
-          ))}
+        <div className="space-y-4">
+          {dayGroups.map((group, gi) => {
+            const filtered = applyFilter(group.recommendations);
+            if (filter !== "all" && filtered.length === 0) return null;
+
+            const isExpanded = expandedDays.has(group.date);
+            const isToday = gi === 0;
+
+            return (
+              <div key={group.date} className="animate-fade-in-up" style={{ animationDelay: `${gi * 60}ms` }}>
+                {/* Date header */}
+                <button
+                  onClick={() => toggleDay(group.date)}
+                  className="w-full flex items-center gap-3 px-4 py-3 glass-card-static hover:bg-white/[0.03] transition-colors group"
+                >
+                  <svg
+                    width="14"
+                    height="14"
+                    viewBox="0 0 24 24"
+                    fill="none"
+                    stroke="currentColor"
+                    strokeWidth="2"
+                    strokeLinecap="round"
+                    strokeLinejoin="round"
+                    className={`text-text-muted/60 transition-transform duration-200 shrink-0 ${isExpanded ? "rotate-90" : ""}`}
+                  >
+                    <path d="M9 18l6-6-6-6" />
+                  </svg>
+
+                  <div className="flex items-center gap-2.5 flex-1 text-left">
+                    <span className={`text-sm font-semibold ${isToday ? "text-amber-400" : "text-text-primary"}`}>
+                      {formatDate(group.date)}
+                    </span>
+                    <span className="text-[10px] text-text-muted/60 font-mono tabular-nums">
+                      {group.date}
+                    </span>
+                    {isToday && (
+                      <span className="text-[10px] px-1.5 py-0.5 rounded-md bg-amber-500/15 text-amber-400 border border-amber-500/20 font-medium">
+                        NEW
+                      </span>
+                    )}
+                  </div>
+
+                  <div className="flex items-center gap-4 text-xs text-text-muted shrink-0">
+                    <div className="flex items-center gap-1.5">
+                      <span className="text-text-muted/50">均分</span>
+                      <span className="font-mono tabular-nums font-semibold text-text-secondary">
+                        {group.avg_score}
+                      </span>
+                    </div>
+                    <div className="flex items-center gap-1.5">
+                      <span className="text-text-muted/50">买入</span>
+                      <span className="font-mono tabular-nums font-semibold text-red-400">
+                        {group.buy_count}
+                      </span>
+                    </div>
+                    <div className="flex items-center gap-1.5">
+                      <span className="text-text-muted/50">共</span>
+                      <span className="font-mono tabular-nums font-semibold text-text-secondary">
+                        {filter === "all" ? group.count : filtered.length}
+                      </span>
+                      <span className="text-text-muted/50">只</span>
+                    </div>
+                  </div>
+                </button>
+
+                {/* Cards */}
+                {isExpanded && (
+                  <div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-3 mt-2 px-1">
+                    {filtered.map((rec, i) => (
+                      <div
+                        key={`${group.date}-${rec.ts_code}`}
+                        className="animate-fade-in-up"
+                        style={{ animationDelay: `${i * 40}ms` }}
+                      >
+                        <StockCard rec={rec} showLLMLoading={isToday && llmEnabled} />
+                      </div>
+                    ))}
+                  </div>
+                )}
+              </div>
+            );
+          })}
         </div>
       )}
     </div>

frontend/src/app/sectors/page.tsx

@@ -2,10 +2,186 @@
 
 import { useEffect, useState, useCallback } from "react";
 import { fetchAPI } from "@/lib/api";
-import type { SectorData } from "@/lib/api";
-import SectorHeatmap from "@/components/sector-heatmap";
+import type { SectorData, LeadingStock } from "@/lib/api";
+import { formatNumber } from "@/lib/utils";
 import { useWebSocket } from "@/hooks/use-websocket";
 
+function getStageInfo(stage: string) {
+  switch (stage) {
+    case "early":
+      return { label: "启动期", color: "text-emerald-400", bg: "bg-emerald-500/10 border-emerald-500/15" };
+    case "mid":
+      return { label: "发展期", color: "text-amber-400", bg: "bg-amber-500/10 border-amber-500/15" };
+    case "late":
+      return { label: "后期", color: "text-orange-400", bg: "bg-orange-500/10 border-orange-500/15" };
+    case "end":
+      return { label: "尾声", color: "text-red-400", bg: "bg-red-500/10 border-red-500/15" };
+    default:
+      return { label: "—", color: "text-text-muted", bg: "bg-white/[0.03] border-white/[0.06]" };
+  }
+}
+
+/** 迷你柱状图：近5日涨跌幅 */
+function MiniBarChart({ data }: { data: number[] }) {
+  if (!data.length) return null;
+  const maxAbs = Math.max(...data.map(Math.abs), 0.1);
+  return (
+    <div className="flex items-end gap-1 h-8">
+      {data.map((v, i) => {
+        const h = Math.max(Math.abs(v) / maxAbs * 100, 8);
+        return (
+          <div key={i} className="flex-1 flex flex-col justify-end items-center" style={{ height: "100%" }}>
+            <div
+              className={`w-full rounded-sm transition-all duration-300 ${v >= 0 ? "bg-red-500/50" : "bg-emerald-500/50"}`}
+              style={{ height: `${h}%` }}
+            />
+          </div>
+        );
+      })}
+    </div>
+  );
+}
+
+/** 领涨股标签 */
+function LeadingStockTag({ stock }: { stock: LeadingStock }) {
+  const isLimitUp = stock.pct_chg >= 9.8;
+  return (
+    <a
+      href={`/stock/${stock.ts_code}`}
+      className="inline-flex items-center gap-1 text-[11px] px-2 py-1 rounded-lg bg-white/[0.03] border border-white/[0.06] hover:bg-white/[0.06] transition-colors"
+    >
+      <span className="text-text-secondary font-medium">{stock.name}</span>
+      <span className={`font-mono tabular-nums ${isLimitUp ? "text-red-400 font-bold" : stock.pct_chg > 0 ? "text-red-400/80" : "text-emerald-400/80"}`}>
+        {stock.pct_chg > 0 ? "+" : ""}{stock.pct_chg}%
+      </span>
+      {stock.limit_times != null && stock.limit_times > 1 && (
+        <span className="text-[9px] px-1 py-0.5 rounded bg-red-500/15 text-red-400 font-bold">
+          {stock.limit_times}连板
+        </span>
+      )}
+    </a>
+  );
+}
+
+function SectorDetailCard({ sector, index }: { sector: SectorData; index: number }) {
+  const displayPct = sector.realtime_pct_change ?? sector.pct_change;
+  const isUp = displayPct > 0;
+  const displayLimitUp = sector.realtime_limit_up_count ?? sector.limit_up_count;
+  const leaders = sector.is_realtime
+    ? (sector.leading_stocks_realtime?.length ? sector.leading_stocks_realtime : sector.leading_stocks)
+    : sector.leading_stocks;
+  const stage = getStageInfo(sector.stage ?? "");
+  const isTop3 = index < 3;
+
+  return (
+    <div
+      className="glass-card p-5 animate-fade-in-up"
+      style={{ animationDelay: `${index * 60}ms` }}
+    >
+      {/* Header row */}
+      <div className="flex items-start justify-between mb-3">
+        <div className="flex items-center gap-2.5">
+          <span className={`w-7 h-7 rounded-lg flex items-center justify-center text-xs font-bold shrink-0 ${
+            index === 0
+              ? "bg-gradient-to-br from-amber-500/30 to-amber-600/20 text-amber-400 border border-amber-500/20"
+              : index === 1
+              ? "bg-gradient-to-br from-slate-400/20 to-slate-500/15 text-slate-300 border border-slate-400/15"
+              : index === 2
+              ? "bg-gradient-to-br from-amber-700/20 to-amber-800/15 text-amber-400 border border-amber-600/15"
+              : "bg-white/[0.03] text-text-muted border border-white/[0.04]"
+          }`}>
+            {index + 1}
+          </span>
+          <div>
+            <div className="flex items-center gap-2">
+              <span className={`text-sm font-semibold ${isTop3 ? "text-text-primary" : "text-text-secondary"}`}>
+                {sector.sector_name}
+              </span>
+              {sector.is_realtime && (
+                <span className="text-[9px] px-1.5 py-0.5 rounded bg-emerald-500/10 text-emerald-400/80 border border-emerald-500/15">
+                  实时
+                </span>
+              )}
+            </div>
+            <div className="text-[11px] text-text-muted/60 mt-0.5">
+              {sector.member_count ? `${sector.member_count}只成分股` : ""}
+              {sector.member_count && sector.turnover_avg ? " · " : ""}
+              {sector.turnover_avg ? `均换手 ${sector.turnover_avg}%` : ""}
+            </div>
+          </div>
+        </div>
+        <div className="text-right">
+          <div className={`text-base font-bold font-mono tabular-nums ${isUp ? "text-red-400" : "text-emerald-400"}`}>
+            {displayPct > 0 ? "+" : ""}{displayPct.toFixed(2)}%
+          </div>
+          <div className="flex items-center gap-2 mt-0.5">
+            <span className={`text-[10px] px-1.5 py-0.5 rounded-md border font-medium ${stage.bg} ${stage.color}`}>
+              {stage.label}
+            </span>
+            <span className="text-[10px] text-text-muted/60">
+              连{sector.days_continuous}天
+            </span>
+          </div>
+        </div>
+      </div>
+
+      {/* Metrics row */}
+      <div className="grid grid-cols-3 gap-3 mb-3">
+        <div className="bg-white/[0.02] rounded-lg px-3 py-2">
+          <div className="text-[10px] text-text-muted/50 mb-0.5">资金净流入</div>
+          <div className={`text-xs font-mono tabular-nums font-semibold ${sector.capital_inflow > 0 ? "text-red-400" : "text-emerald-400"}`}>
+            {sector.capital_inflow > 0 ? "+" : ""}{formatNumber(sector.capital_inflow)}
+          </div>
+        </div>
+        <div className="bg-white/[0.02] rounded-lg px-3 py-2">
+          <div className="text-[10px] text-text-muted/50 mb-0.5">涨停股</div>
+          <div className="text-xs font-mono tabular-nums font-semibold text-text-secondary">
+            {displayLimitUp}<span className="text-text-muted/40"> 只</span>
+          </div>
+        </div>
+        <div className="bg-white/[0.02] rounded-lg px-3 py-2">
+          <div className="text-[10px] text-text-muted/50 mb-0.5">热度评分</div>
+          <div className={`text-xs font-mono tabular-nums font-semibold ${
+            sector.heat_score >= 70 ? "text-amber-400" : sector.heat_score >= 50 ? "text-text-secondary" : "text-text-muted"
+          }`}>
+            {sector.heat_score.toFixed(0)}
+            <span className="text-text-muted/40 text-[10px]">/100</span>
+          </div>
+        </div>
+      </div>
+
+      {/* 5日趋势图 */}
+      {sector.pct_trend && sector.pct_trend.length > 1 && (
+        <div className="mb-3">
+          <div className="flex items-center justify-between mb-1.5">
+            <span className="text-[10px] text-text-muted/50">近5日走势</span>
+            <div className="flex gap-2">
+              {sector.pct_trend.map((v, i) => (
+                <span key={i} className={`text-[9px] font-mono tabular-nums ${v >= 0 ? "text-red-400/60" : "text-emerald-400/60"}`}>
+                  {v > 0 ? "+" : ""}{v.toFixed(1)}
+                </span>
+              ))}
+            </div>
+          </div>
+          <MiniBarChart data={sector.pct_trend} />
+        </div>
+      )}
+
+      {/* 领涨股 */}
+      {leaders && leaders.length > 0 && (
+        <div>
+          <span className="text-[10px] text-text-muted/50 mb-1.5 block">领涨股</span>
+          <div className="flex flex-wrap gap-1.5">
+            {leaders.map((s) => (
+              <LeadingStockTag key={s.ts_code} stock={s} />
+            ))}
+          </div>
+        </div>
+      )}
+    </div>
+  );
+}
+
 export default function SectorsPage() {
   const [sectors, setSectors] = useState<SectorData[]>([]);
 
@@ -22,20 +198,38 @@ export default function SectorsPage() {
     loadData();
   }, [loadData]);
 
-  // WebSocket 触发刷新（扫描完成时更新）
   useWebSocket(
     useCallback(() => {
       loadData();
     }, [loadData])
   );
 
+  const hasRealtime = sectors.some((s) => s.is_realtime);
+
   return (
-    <div className="max-w-5xl mx-auto px-4 md:px-8 pt-6 pb-20 md:pb-10">
-      <div className="mb-5 animate-fade-in-up">
-        <h1 className="text-lg font-bold tracking-tight">板块分析</h1>
-        <p className="text-xs text-text-muted mt-0.5">基于资金流向和涨跌表现的热度排名</p>
+    <div className="max-w-7xl mx-auto px-4 md:px-8 pt-6 pb-20 md:pb-10">
+      <div className="flex items-center justify-between mb-5 animate-fade-in-up">
+        <div>
+          <h1 className="text-lg font-bold tracking-tight">板块分析</h1>
+          <p className="text-xs text-text-muted mt-0.5">
+            基于资金流向、涨跌表现、龙头股的热度排名
+            {hasRealtime && <span className="text-emerald-400/60 ml-1">· 实时数据</span>}
+          </p>
+        </div>
       </div>
-      <SectorHeatmap sectors={sectors} />
+
+      {!sectors.length ? (
+        <div className="glass-card-static p-12 text-center animate-fade-in-up">
+          <div className="text-text-muted text-sm mb-1">暂无板块数据</div>
+          <div className="text-text-muted/50 text-xs">触发扫描后自动更新</div>
+        </div>
+      ) : (
+        <div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-4">
+          {sectors.map((sector, i) => (
+            <SectorDetailCard key={sector.sector_code} sector={sector} index={i} />
+          ))}
+        </div>
+      )}
     </div>
   );
 }

frontend/src/components/stock-card.tsx

@@ -1,16 +1,27 @@
 "use client";
 
+import { useState } from "react";
 import { getLevelBadge, getSignalColor, getScoreColor } from "@/lib/utils";
 import type { RecommendationData } from "@/lib/api";
 
 export default function StockCard({ rec, showLLMLoading = false }: { rec: RecommendationData; showLLMLoading?: boolean }) {
   const badge = getLevelBadge(rec.level);
+  const [aiExpanded, setAiExpanded] = useState(false);
 
-  // 策略标签
-  const strategyLabel = rec.strategy === "potential" ? "潜在启动" : "强中选强";
-  const strategyStyle = rec.strategy === "potential"
-    ? "bg-cyan-500/15 text-cyan-400 border-cyan-500/20"
-    : "bg-amber-500/15 text-amber-400 border-amber-500/20";
+  // 入场信号标签
+  const signalTypeMap: Record<string, { label: string; style: string }> = {
+    breakout: { label: "突破型", style: "bg-red-500/15 text-red-400 border-red-500/20" },
+    pullback: { label: "回踩型", style: "bg-blue-500/15 text-blue-400 border-blue-500/20" },
+    launch: { label: "启动型", style: "bg-orange-500/15 text-orange-400 border-orange-500/20" },
+  };
+  // 向后兼容：旧数据使用 strategy 字段
+  const signalInfo = signalTypeMap[rec.entry_signal_type || ""];
+  const legacyStrategy = rec.strategy === "potential"
+    ? { label: "潜在启动", style: "bg-cyan-500/15 text-cyan-400 border-cyan-500/20" }
+    : rec.strategy === "momentum"
+    ? { label: "强中选强", style: "bg-amber-500/15 text-amber-400 border-amber-500/20" }
+    : null;
+  const tag = signalInfo || legacyStrategy;
 
   return (
     <a
@@ -25,9 +36,11 @@ export default function StockCard({ rec, showLLMLoading = false }: { rec: Recomm
             <span className={`text-xs px-2 py-0.5 rounded-full font-medium ${badge.bg} ${badge.text}`}>
               {rec.level}
             </span>
-            <span className={`text-[10px] px-1.5 py-0.5 rounded-md font-medium border ${strategyStyle}`}>
-              {strategyLabel}
-            </span>
+            {tag && (
+              <span className={`text-[10px] px-1.5 py-0.5 rounded-md font-medium border ${tag.style}`}>
+                {tag.label}
+              </span>
+            )}
           </div>
           <div className="text-xs text-text-muted mt-1 font-mono tabular-nums">
             {rec.ts_code} <span className="text-text-muted/40 mx-1">·</span> {rec.sector}
@@ -79,24 +92,34 @@ export default function StockCard({ rec, showLLMLoading = false }: { rec: Recomm
         ))}
       </div>
 
-      {/* AI Analysis */}
+      {/* AI Analysis — collapsible */}
       {rec.llm_analysis && rec.llm_analysis !== "AI 分析暂时不可用" ? (
-        <div className="mt-3 bg-accent-cyan/[0.06] border border-accent-cyan/[0.12] rounded-xl px-4 py-3">
-          <div className="flex items-center justify-between mb-1.5">
-            <div className="text-xs text-accent-cyan/80 font-semibold tracking-wider">AI 分析</div>
+        <div className="mt-3">
+          <button
+            onClick={(e) => { e.preventDefault(); setAiExpanded(!aiExpanded); }}
+            className="w-full flex items-center justify-between bg-accent-cyan/[0.06] border border-accent-cyan/[0.12] rounded-xl px-4 py-2.5 hover:bg-accent-cyan/[0.09] transition-colors"
+          >
+            <div className="flex items-center gap-2 text-xs text-accent-cyan/80 font-semibold tracking-wider">
+              <svg width="12" height="12" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="2" strokeLinecap="round" strokeLinejoin="round" className={`transition-transform duration-200 ${aiExpanded ? "rotate-90" : ""}`}>
+                <path d="M9 18l6-6-6-6" />
+              </svg>
+              AI 分析
+            </div>
             {rec.llm_score != null && (
               <div className="text-xs font-mono tabular-nums">
-                <span className="text-text-muted">AI评分 </span>
+                <span className="text-text-muted">评分 </span>
                 <span className={`font-bold ${rec.llm_score >= 7 ? "text-amber-400" : rec.llm_score >= 5 ? "text-cyan-400" : "text-text-muted"}`}>
                   {rec.llm_score}
                 </span>
                 <span className="text-text-muted/50">/10</span>
               </div>
             )}
-          </div>
-          <div className="text-xs text-text-secondary leading-relaxed whitespace-pre-line">
-            <MarkdownText text={rec.llm_analysis} />
-          </div>
+          </button>
+          {aiExpanded && (
+            <div className="text-xs text-text-secondary leading-relaxed whitespace-pre-line bg-accent-cyan/[0.03] border border-t-0 border-accent-cyan/[0.08] rounded-b-xl px-4 py-3">
+              <MarkdownText text={rec.llm_analysis} />
+            </div>
+          )}
         </div>
       ) : rec.llm_analysis === "AI 分析暂时不可用" ? (
         <div className="mt-3 text-xs text-text-muted/50 flex items-center gap-1.5">

frontend/src/lib/api.ts

@@ -109,13 +109,22 @@ export interface RecommendationData {
   stop_loss: number | null;
   reasons: string[];
   risk_note: string;
-  strategy?: "momentum" | "potential";
+  strategy?: "momentum" | "potential" | "trend_breakout";
+  entry_signal_type?: "breakout" | "pullback" | "launch" | "none";
   llm_analysis?: string;
   llm_score?: number | null;
   scan_session: string;
   created_at: string | null;
 }
 
+export interface LeadingStock {
+  ts_code: string;
+  name: string;
+  pct_chg: number;
+  amount: number;
+  limit_times?: number;
+}
+
 export interface SectorData {
   sector_code: string;
   sector_name: string;
@@ -124,6 +133,13 @@ export interface SectorData {
   limit_up_count: number;
   days_continuous: number;
   heat_score: number;
+  stage?: string;
+  member_count?: number;
+  leading_stocks?: LeadingStock[];
+  leading_stocks_realtime?: LeadingStock[] | null;
+  pct_trend?: number[];
+  turnover_avg?: number;
+  main_force_ratio?: number;
   realtime_pct_change?: number | null;
   realtime_limit_up_count?: number | null;
   is_realtime?: boolean;
@@ -134,6 +150,14 @@ export interface LatestResult {
   recommendations: RecommendationData[];
 }
 
+export interface DayGroup {
+  date: string;
+  count: number;
+  buy_count: number;
+  avg_score: number;
+  recommendations: RecommendationData[];
+}
+
 export interface ChatMessage {
   role: "user" | "assistant";
   content: string;