diff --git a/backend/.env b/backend/.env
index 1ee376ed..dac3bc55 100644
--- a/backend/.env
+++ b/backend/.env
@@ -1,7 +1,7 @@
 ASTOCK_TUSHARE_TOKEN=0ed6419a00d8923dc19c0b58fc92d94c9a0696949ab91a13aa58a0cc
 ASTOCK_DEBUG=true
 
-ASTOCK_DEEPSEEK_API_KEY=sk-9f6b56f08796435d988cf202e37f6ee3
+ASTOCK_DEEPSEEK_API_KEY=sk-ee8eee63d5cf41eba14a328de49055ac
 ASTOCK_ALERT_ENABLED=true
 ASTOCK_FEISHU_WEBHOOK_URL=https://open.feishu.cn/open-apis/bot/v2/hook/6307668f-10aa-4fc1-8c1e-bad1b6b78d4d
 ASTOCK_ALERT_ENVIRONMENT=local
diff --git a/backend/app/api/catalysts.py b/backend/app/api/catalysts.py
index 9d7f51f6..33b6f773 100644
--- a/backend/app/api/catalysts.py
+++ b/backend/app/api/catalysts.py
@@ -3,8 +3,14 @@
 from fastapi import APIRouter, Depends
 
 from app.catalyst.models import CatalystInput
-from app.catalyst.service import build_theme_catalyst_scores, get_recent_catalysts, ingest_catalyst
+from app.catalyst.service import (
+    build_theme_catalyst_scores,
+    get_recent_catalysts,
+    get_recent_news_items,
+    ingest_catalyst,
+)
 from app.core.deps import get_current_admin
+from app.news.pipeline import refresh_news_catalysts
 
 router = APIRouter(prefix="/api/catalysts", tags=["catalysts"])
 
@@ -14,6 +20,11 @@ async def recent(limit: int = 30, hours: int = 72):
     return await get_recent_catalysts(limit=limit, hours=hours)
 
 
+@router.get("/news")
+async def news(limit: int = 50, hours: int = 24, status: str | None = None):
+    return await get_recent_news_items(limit=limit, hours=hours, status=status)
+
+
 @router.get("/theme-scores")
 async def theme_scores(hours: int = 72, limit: int = 20):
     scores = await build_theme_catalyst_scores(hours=hours, limit=limit)
@@ -24,3 +35,8 @@ async def theme_scores(hours: int = 72, limit: int = 20):
 async def ingest(item: CatalystInput, _admin: dict = Depends(get_current_admin)):
     analysis = await ingest_catalyst(item, use_llm=True)
     return analysis.model_dump()
+
+
+@router.post("/refresh-news")
+async def refresh_news(_admin: dict = Depends(get_current_admin)):
+    return await refresh_news_catalysts()
diff --git a/backend/app/api/stocks.py b/backend/app/api/stocks.py
index 0eed9601..b5f688e6 100644
--- a/backend/app/api/stocks.py
+++ b/backend/app/api/stocks.py
@@ -10,6 +10,7 @@ from starlette.responses import StreamingResponse
 
 from app.data.tushare_client import tushare_client
 from app.data import tencent_client
+from app.data.code_utils import normalize_ts_code
 from app.analysis.technical import add_all_indicators
 from app.analysis.signals import generate_signals
 from app.db.database import get_db
@@ -38,6 +39,7 @@ async def get_stock_thesis(ts_code: str):
     """获取个股推荐推演归档（只读缓存，不触发扫描或 LLM）。"""
     from sqlalchemy import text
 
+    ts_code = normalize_ts_code(ts_code)
     async with get_db() as db:
         rec_result = await db.execute(
             text(
@@ -139,7 +141,6 @@ async def get_stock_thesis(ts_code: str):
     decision_points = [
         {"label": "操作计划", "value": r["action_plan"] or "观察"},
         {"label": "召回来源", "value": " / ".join(_safe_json_list(r.get("recall_tags"))) or "未归档"},
-        {"label": "AI预筛", "value": r.get("prefilter_decision") or "未执行"},
         {"label": "触发条件", "value": r["trigger_condition"] or "等待触发条件归档"},
         {"label": "失效条件", "value": r["invalidation_condition"] or "等待失效条件归档"},
         {"label": "建议仓位", "value": f"{r['suggested_position_pct']}%" if r["suggested_position_pct"] is not None else "未设置"},
@@ -220,6 +221,7 @@ async def get_stock_thesis(ts_code: str):
 @router.get("/{ts_code}/quote")
 async def get_quote(ts_code: str):
     """获取个股实时行情"""
+    ts_code = normalize_ts_code(ts_code)
     quote = await tencent_client.get_realtime_quote(ts_code)
     if not quote:
         return {"error": "获取行情失败"}
@@ -229,6 +231,7 @@ async def get_quote(ts_code: str):
 @router.get("/{ts_code}/kline")
 async def get_kline(ts_code: str, days: int = 120):
     """获取个股K线数据（含技术指标）"""
+    ts_code = normalize_ts_code(ts_code)
     df = tushare_client.get_stock_daily(ts_code, days=days)
     if df.empty:
         return []
@@ -247,6 +250,7 @@ async def get_kline(ts_code: str, days: int = 120):
 @router.get("/{ts_code}/signals")
 async def get_signals(ts_code: str):
     """获取个股技术面买卖信号"""
+    ts_code = normalize_ts_code(ts_code)
     signal = generate_signals(ts_code)
     return signal.model_dump()
 
@@ -254,6 +258,7 @@ async def get_signals(ts_code: str):
 @router.get("/{ts_code}/capital_flow")
 async def get_capital_flow(ts_code: str, days: int = 10):
     """获取个股资金流向（含大/中/小单分拆）"""
+    ts_code = normalize_ts_code(ts_code)
     df = tushare_client.get_stock_moneyflow(ts_code, days=days)
     if df.empty:
         return []
@@ -287,6 +292,7 @@ async def get_capital_flow(ts_code: str, days: int = 10):
 @router.get("/{ts_code}/diagnose/history")
 async def get_diagnose_history(ts_code: str):
     """获取个股最近5次诊断历史"""
+    ts_code = normalize_ts_code(ts_code)
     try:
         from sqlalchemy import text
         async with get_db() as db:
@@ -321,6 +327,7 @@ async def get_diagnose_history(ts_code: str):
 @router.post("/{ts_code}/diagnose")
 async def diagnose_stock(ts_code: str, mode: str = Query("entry")):
     """AI 诊断个股（SSE 流式返回）"""
+    ts_code = normalize_ts_code(ts_code)
     from app.config import settings
     if not settings.deepseek_api_key:
         return {"status": "error", "message": "未配置 LLM API Key"}
@@ -484,7 +491,7 @@ async def diagnose_stock(ts_code: str, mode: str = Query("entry")):
                     "capital_score, technical_score, position_score, sector, signal, entry_signal_type "
                     "FROM recommendations "
                     "WHERE ts_code = :code "
-                    "AND (action_plan IN ('可操作', '重点关注') OR COALESCE(llm_score, 0) >= 6 OR score >= 56) "
+                    "AND (action_plan IN ('可操作', '重点关注') OR score >= 56) "
                     "ORDER BY created_at DESC LIMIT 1"
                 ),
                 {"code": ts_code},
diff --git a/backend/app/catalyst/service.py b/backend/app/catalyst/service.py
index b1791f16..31e12bd8 100644
--- a/backend/app/catalyst/service.py
+++ b/backend/app/catalyst/service.py
@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import hashlib
 import logging
 from datetime import datetime, timedelta
 
@@ -11,6 +12,7 @@ from app.catalyst.mapper import analyze_catalyst
 from app.catalyst.models import CatalystAnalysis, CatalystInput, ThemeCatalystScore
 from app.db import tables
 from app.db.database import get_db
+from app.news.models import NewsItem
 
 logger = logging.getLogger(__name__)
 
@@ -21,8 +23,26 @@ async def ingest_catalyst(item: CatalystInput, use_llm: bool = True) -> Catalyst
     return analysis
 
 
+async def ingest_catalyst_with_id(item: CatalystInput, use_llm: bool = True) -> tuple[CatalystAnalysis, int]:
+    analysis = await analyze_catalyst(item, use_llm=use_llm)
+    catalyst_id = await save_catalyst(analysis)
+    return analysis, catalyst_id
+
+
 async def save_catalyst(analysis: CatalystAnalysis) -> int:
     async with get_db() as db:
+        if analysis.url:
+            exists = await db.execute(
+                text(
+                    "SELECT id FROM catalysts WHERE source = :source AND url = :url "
+                    "ORDER BY id DESC LIMIT 1"
+                ),
+                {"source": analysis.source, "url": analysis.url},
+            )
+            row = exists.fetchone()
+            if row:
+                return int(row._mapping["id"])
+
         result = await db.execute(
             tables.catalysts_table.insert().values(
                 title=analysis.title,
@@ -58,6 +78,146 @@ async def save_catalyst(analysis: CatalystAnalysis) -> int:
         return catalyst_id
 
 
+async def ingest_news_items(items: list[NewsItem]) -> dict:
+    """保存原始新闻并分析新增项。"""
+    inserted = 0
+    duplicates = 0
+    analyzed = 0
+    failed = 0
+
+    async with get_db() as db:
+        for item in items:
+            dedup_key = build_news_dedup_key(item)
+            exists = await db.execute(
+                text("SELECT id FROM news_items WHERE dedup_key = :dedup_key LIMIT 1"),
+                {"dedup_key": dedup_key},
+            )
+            if exists.fetchone():
+                duplicates += 1
+                continue
+
+            await db.execute(
+                tables.news_items_table.insert().values(
+                    title=item.title,
+                    content=item.content,
+                    summary=item.summary,
+                    source=item.source,
+                    url=item.url,
+                    published_at=item.published_at,
+                    dedup_key=dedup_key,
+                    status="pending",
+                )
+            )
+            inserted += 1
+        await db.commit()
+
+    if inserted:
+        result = await analyze_pending_news(limit=inserted)
+        analyzed += int(result.get("analyzed", 0))
+        failed += int(result.get("failed", 0))
+
+    return {
+        "fetched": len(items),
+        "inserted": inserted,
+        "duplicates": duplicates,
+        "analyzed": analyzed,
+        "failed": failed,
+    }
+
+
+async def analyze_pending_news(limit: int = 50, use_llm: bool = True) -> dict:
+    rows = []
+    async with get_db() as db:
+        result = await db.execute(
+            text(
+                "SELECT * FROM news_items "
+                "WHERE status = 'pending' "
+                "ORDER BY COALESCE(published_at, created_at) DESC, id DESC "
+                "LIMIT :limit"
+            ),
+            {"limit": limit},
+        )
+        rows = [dict(row._mapping) for row in result.fetchall()]
+
+    analyzed = 0
+    skipped = 0
+    failed = 0
+
+    for row in rows:
+        title = row.get("title") or ""
+        content = row.get("content") or row.get("summary") or ""
+        try:
+            if not _looks_market_relevant(title, content):
+                await _mark_news_item(row["id"], status="skipped", error="")
+                skipped += 1
+                continue
+
+            _, catalyst_id = await ingest_catalyst_with_id(
+                CatalystInput(
+                    title=title,
+                    content=content,
+                    source=row.get("source") or "news",
+                    url=row.get("url") or "",
+                    published_at=row.get("published_at"),
+                ),
+                use_llm=use_llm,
+            )
+            await _mark_news_item(
+                row["id"],
+                status="analyzed",
+                catalyst_id=catalyst_id,
+                error="",
+            )
+            analyzed += 1
+        except Exception as e:
+            logger.warning("新闻催化分析失败 id=%s title=%s error=%s", row.get("id"), title, e)
+            await _mark_news_item(row["id"], status="failed", error=str(e)[:500])
+            failed += 1
+
+    return {"analyzed": analyzed, "skipped": skipped, "failed": failed}
+
+
+def build_news_dedup_key(item: NewsItem) -> str:
+    text = f"{item.source}|{item.url or item.title}"
+    normalized = "".join(ch.lower() for ch in text.strip() if ch.isalnum() or ch in ".:/_-|")
+    return hashlib.sha256(normalized.encode("utf-8")).hexdigest()
+
+
+async def _mark_news_item(
+    news_id: int,
+    status: str,
+    catalyst_id: int | None = None,
+    error: str = "",
+) -> None:
+    async with get_db() as db:
+        await db.execute(
+            text(
+                "UPDATE news_items SET status = :status, catalyst_id = :catalyst_id, "
+                "error = :error, analyzed_at = :analyzed_at WHERE id = :id"
+            ),
+            {
+                "status": status,
+                "catalyst_id": catalyst_id,
+                "error": error,
+                "analyzed_at": datetime.now(),
+                "id": news_id,
+            },
+        )
+        await db.commit()
+
+
+def _looks_market_relevant(title: str, content: str) -> bool:
+    text_value = f"{title} {content}"
+    keywords = [
+        "A股", "股市", "上市公司", "证券", "券商", "板块", "行业", "概念",
+        "政策", "国务院", "发改委", "工信部", "央行", "证监会", "交易所",
+        "业绩", "订单", "签约", "并购", "重组", "回购", "定增", "涨停",
+        "资金", "主力", "北向", "算力", "人工智能", "半导体", "新能源",
+        "机器人", "低空", "医药", "消费", "军工", "地产", "有色",
+    ]
+    return any(keyword in text_value for keyword in keywords)
+
+
 async def get_recent_catalysts(limit: int = 30, hours: int = 72) -> list[dict]:
     since = datetime.now() - timedelta(hours=hours)
     async with get_db() as db:
@@ -78,6 +238,27 @@ async def get_recent_catalysts(limit: int = 30, hours: int = 72) -> list[dict]:
     return [dict(row) for row in rows]
 
 
+async def get_recent_news_items(limit: int = 50, hours: int = 24, status: str | None = None) -> list[dict]:
+    since = datetime.now() - timedelta(hours=hours)
+    conditions = ["COALESCE(published_at, created_at) >= :since"]
+    params = {"since": since, "limit": limit}
+    if status:
+        conditions.append("status = :status")
+        params["status"] = status
+
+    async with get_db() as db:
+        result = await db.execute(
+            text(
+                "SELECT id, title, source, url, published_at, status, catalyst_id, error, created_at, analyzed_at "
+                "FROM news_items WHERE "
+                + " AND ".join(conditions)
+                + " ORDER BY COALESCE(published_at, created_at) DESC, id DESC LIMIT :limit"
+            ),
+            params,
+        )
+        return [dict(row._mapping) for row in result.fetchall()]
+
+
 async def build_theme_catalyst_scores(hours: int = 72, limit: int = 20) -> list[ThemeCatalystScore]:
     since = datetime.now() - timedelta(hours=hours)
     async with get_db() as db:
diff --git a/backend/app/config.py b/backend/app/config.py
index 27edf2c0..9a6985de 100644
--- a/backend/app/config.py
+++ b/backend/app/config.py
@@ -31,9 +31,6 @@ class Settings(BaseSettings):
     top_sector_count: int = 5           # 关注板块数量
     top_stock_count: int = 6            # 最终推荐输出上限
     candidate_pool_limit: int = 90      # 多路召回后的候选池上限
-    llm_prefilter_limit: int = 24       # LLM 初筛保留数量
-    llm_prefilter_max_concurrent: int = 6
-    llm_final_limit: int = 10           # LLM 深裁决池上限
     actionable_limit: int = 3           # 最多可操作标的
     watch_limit: int = 5                # 最多重点关注标的
     min_turnover_rate: float = 2.0      # 最小换手率 %
@@ -54,6 +51,16 @@ class Settings(BaseSettings):
     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 %
 
+    # 新闻/政策催化采集
+    news_collection_enabled: bool = True
+    news_tushare_sources: str = "sina,eastmoney,10jqka,wallstreetcn"
+    news_tushare_sources_per_run: int = 1
+    news_rss_sources: str = ""          # name|url,name|url
+    news_fetch_lookback_hours: int = 24
+    news_fetch_limit_per_source: int = 30
+    news_analyze_limit_per_run: int = 50
+    news_min_title_length: int = 8
+
     # LLM (DeepSeek)
     deepseek_api_key: str = ""
     deepseek_base_url: str = "https://api.deepseek.com/v1"
diff --git a/backend/app/data/code_utils.py b/backend/app/data/code_utils.py
new file mode 100644
index 00000000..8654fe2a
--- /dev/null
+++ b/backend/app/data/code_utils.py
@@ -0,0 +1,54 @@
+"""股票代码规范化工具。"""
+
+from __future__ import annotations
+
+import re
+
+SUPPORTED_MARKETS = {"SH", "SZ", "BJ"}
+
+
+def normalize_ts_code(value: str | None) -> str:
+    """把常见 A 股代码输入规范成 Tushare 格式。
+
+    支持:
+      - 603779 -> 603779.SH
+      - 300750 -> 300750.SZ
+      - 430047 -> 430047.BJ
+      - sh600519 / sz000001 / bj430047
+      - 600519.sh -> 600519.SH
+    """
+    text = str(value or "").strip().upper()
+    if not text:
+        return ""
+
+    compact_match = re.fullmatch(r"(SH|SZ|BJ)(\d{6})", text)
+    if compact_match:
+        market, code = compact_match.groups()
+        return f"{code}.{market}"
+
+    dotted_match = re.fullmatch(r"(\d{6})\.(SH|SZ|BJ)", text)
+    if dotted_match:
+        code, market = dotted_match.groups()
+        return f"{code}.{market}"
+
+    if re.fullmatch(r"\d{6}", text):
+        if text.startswith(("6", "9")):
+            return f"{text}.SH"
+        if text.startswith(("0", "2", "3")):
+            return f"{text}.SZ"
+        if text.startswith(("4", "8")):
+            return f"{text}.BJ"
+
+    return text
+
+
+def split_ts_code(value: str | None) -> tuple[str, str]:
+    """返回 (code, market)，无法识别时抛出带上下文的 ValueError。"""
+    normalized = normalize_ts_code(value)
+    if "." not in normalized:
+        raise ValueError(f"无效股票代码格式: {value!r}")
+
+    code, market = normalized.split(".", 1)
+    if not re.fullmatch(r"\d{6}", code) or market not in SUPPORTED_MARKETS:
+        raise ValueError(f"无效股票代码格式: {value!r}")
+    return code, market
diff --git a/backend/app/data/eastmoney_client.py b/backend/app/data/eastmoney_client.py
index 7df4271b..9e9d30e9 100644
--- a/backend/app/data/eastmoney_client.py
+++ b/backend/app/data/eastmoney_client.py
@@ -17,6 +17,7 @@ from datetime import datetime
 
 from app.data.cache import cache
 from app.config import settings
+from app.data.code_utils import split_ts_code
 from app.db.error_logger import log_error
 
 logger = logging.getLogger(__name__)
@@ -57,8 +58,8 @@ def _mark_rows_status(rows: list[dict], status: str, detail: str) -> list[dict]:
 
 
 def _ts_code_to_eastmoney(ts_code: str) -> str:
-    """600519.SH -> 1.600519  (上海=1, 深圳=0)"""
-    code, market = ts_code.split(".")
+    """600519.SH -> 1.600519  (上海=1, 深圳/北交所=0)"""
+    code, market = split_ts_code(ts_code)
     prefix = "1" if market == "SH" else "0"
     return f"{prefix}.{code}"
 
diff --git a/backend/app/data/sina_client.py b/backend/app/data/sina_client.py
index 00b66824..b6a89db9 100644
--- a/backend/app/data/sina_client.py
+++ b/backend/app/data/sina_client.py
@@ -16,6 +16,7 @@ import httpx
 import pandas as pd
 
 from app.data.cache import cache
+from app.data.code_utils import split_ts_code
 from app.data.models import StockQuote
 
 logger = logging.getLogger(__name__)
@@ -29,7 +30,7 @@ HEADERS = {
 
 
 def _ts_code_to_sina(ts_code: str) -> str:
-    code, market = ts_code.split(".")
+    code, market = split_ts_code(ts_code)
     return f"{market.lower()}{code}"
 
 
diff --git a/backend/app/data/tencent_client.py b/backend/app/data/tencent_client.py
index 8b990d00..f28ad34a 100644
--- a/backend/app/data/tencent_client.py
+++ b/backend/app/data/tencent_client.py
@@ -8,6 +8,7 @@ import logging
 import httpx
 from app.data.cache import cache
 from app.config import settings
+from app.data.code_utils import normalize_ts_code, split_ts_code
 from app.data.models import StockQuote
 from app.db.error_logger import log_error
 
@@ -22,7 +23,7 @@ HEADERS = {
 
 def _ts_code_to_tencent(ts_code: str) -> str:
     """600519.SH -> sh600519"""
-    code, market = ts_code.split(".")
+    code, market = split_ts_code(ts_code)
     return f"{market.lower()}{code}"
 
 
@@ -67,6 +68,7 @@ def _parse_tencent_response(raw: str) -> dict | None:
 
 async def get_realtime_quote(ts_code: str) -> StockQuote | None:
     """获取单只股票实时行情"""
+    ts_code = normalize_ts_code(ts_code)
     cache_key = f"rt_quote:{ts_code}"
     cached = cache.get(cache_key)
     if cached is not None:
@@ -118,6 +120,7 @@ async def get_realtime_quotes_batch(ts_codes: list[str]) -> dict[str, StockQuote
     """批量获取实时行情（腾讯支持逗号拼接多只）"""
     results = {}
     uncached = []
+    ts_codes = [normalize_ts_code(code) for code in ts_codes if normalize_ts_code(code)]
 
     for code in ts_codes:
         cached = cache.get(f"rt_quote:{code}")
@@ -191,6 +194,7 @@ async def get_index_realtime(index_codes: list[str] = None) -> dict[str, dict]:
     """
     if not index_codes:
         index_codes = ["000001.SH", "399001.SZ", "399006.SZ"]
+    index_codes = [normalize_ts_code(code) for code in index_codes if normalize_ts_code(code)]
 
     results = {}
     symbols = ",".join(_ts_code_to_tencent(c) for c in index_codes)
diff --git a/backend/app/data/tushare_client.py b/backend/app/data/tushare_client.py
index 178e07c1..f2218e56 100644
--- a/backend/app/data/tushare_client.py
+++ b/backend/app/data/tushare_client.py
@@ -265,5 +265,43 @@ class TushareClient:
             )
         )
 
+    # ── 新闻快讯 ──
+
+    def get_news(
+        self,
+        source: str,
+        start_time: datetime,
+        end_time: datetime,
+        limit: int = 30,
+    ) -> pd.DataFrame:
+        """获取 Tushare 新闻快讯。
+
+        Tushare news 接口不同账号权限不一，失败时返回空 DataFrame，
+        不阻断其他新闻源或推荐流程。该接口常见限频为 1 次/分钟或
+        2 次/小时，因此这里不复用通用重试逻辑，避免失败重试继续消耗配额。
+        """
+        cache_key = f"news:{source}:{start_time:%Y%m%d%H}:{end_time:%Y%m%d%H}:{limit}"
+        cached = cache.get(cache_key)
+        if cached is not None:
+            return cached
+
+        try:
+            self._ensure_init()
+            self._rate_limit()
+            result = self.pro.news(
+                src=source,
+                start_date=start_time.strftime("%Y-%m-%d %H:%M:%S"),
+                end_date=end_time.strftime("%Y-%m-%d %H:%M:%S"),
+                limit=limit,
+            )
+            if result is None or result.empty:
+                return pd.DataFrame()
+            cache.set(cache_key, result, 600)
+            return result
+        except Exception as e:
+            logger.warning("Tushare 新闻请求失败 source=%s: %s", source, e)
+            log_error_background("tushare_news", f"Tushare 新闻请求失败 source={source}: {e}")
+            return pd.DataFrame()
+
 
 tushare_client = TushareClient()
diff --git a/backend/app/db/database.py b/backend/app/db/database.py
index 2527baec..5c00bcd2 100644
--- a/backend/app/db/database.py
+++ b/backend/app/db/database.py
@@ -98,6 +98,8 @@ async def init_db():
             "ALTER TABLE strategy_configs ADD COLUMN effective_from DATETIME DEFAULT CURRENT_TIMESTAMP",
             "ALTER TABLE prompt_configs ADD COLUMN evidence_json TEXT DEFAULT '{}'",
             "ALTER TABLE strategy_config_changes ADD COLUMN prompt_key TEXT DEFAULT ''",
+            "ALTER TABLE news_items ADD COLUMN summary TEXT DEFAULT ''",
+            "ALTER TABLE news_items ADD COLUMN error TEXT DEFAULT ''",
             "ALTER TABLE catalysts ADD COLUMN llm_reason TEXT DEFAULT ''",
         ]:
             try:
@@ -107,6 +109,16 @@ async def init_db():
             except Exception:
                 pass  # 列已存在，忽略
 
+        for index_sql in [
+            "CREATE UNIQUE INDEX IF NOT EXISTS idx_news_items_dedup_key ON news_items(dedup_key)",
+            "CREATE INDEX IF NOT EXISTS idx_news_items_status_time ON news_items(status, published_at)",
+            "CREATE INDEX IF NOT EXISTS idx_catalysts_source_url ON catalysts(source, url)",
+        ]:
+            try:
+                await conn.execute(__import__("sqlalchemy").text(index_sql))
+            except Exception:
+                pass
+
         try:
             await conn.execute(
                 __import__("sqlalchemy").text(
diff --git a/backend/app/db/tables.py b/backend/app/db/tables.py
index d28cbaca..1245dcd3 100644
--- a/backend/app/db/tables.py
+++ b/backend/app/db/tables.py
@@ -240,6 +240,23 @@ strategy_config_changes_table = Table(
     Column("applied_at", DateTime),
 )
 
+news_items_table = Table(
+    "news_items", metadata,
+    Column("id", Integer, primary_key=True, autoincrement=True),
+    Column("title", Text, nullable=False),
+    Column("content", Text, default=""),
+    Column("summary", Text, default=""),
+    Column("source", Text, default=""),
+    Column("url", Text, default=""),
+    Column("published_at", DateTime),
+    Column("dedup_key", Text, nullable=False, unique=True),
+    Column("status", Text, default="pending"),  # pending / analyzed / skipped / failed
+    Column("catalyst_id", Integer),
+    Column("error", Text, default=""),
+    Column("created_at", DateTime, server_default=func.now()),
+    Column("analyzed_at", DateTime),
+)
+
 catalysts_table = Table(
     "catalysts", metadata,
     Column("id", Integer, primary_key=True, autoincrement=True),
diff --git a/backend/app/engine/recommender.py b/backend/app/engine/recommender.py
index f1968389..e695ee97 100644
--- a/backend/app/engine/recommender.py
+++ b/backend/app/engine/recommender.py
@@ -577,7 +577,7 @@ async def get_recommendation_history(days: int = 7) -> list[dict]:
             "  ) lt ON t.id = lt.max_id"
             ") latest_t ON latest_t.recommendation_id = r.id "
             "WHERE r.created_at >= :start "
-            "AND (r.action_plan IN ('可操作', '重点关注') OR COALESCE(r.llm_score, 0) >= 6 OR r.score >= 56) "
+            "AND (r.action_plan IN ('可操作', '重点关注') OR r.score >= 56) "
             "AND ("
             "  COALESCE(r.recall_tags, '[]') LIKE '%hot_theme_core%' "
             "  OR COALESCE(r.recall_tags, '[]') LIKE '%theme_leader%' "
@@ -810,7 +810,6 @@ async def _save_to_db(result: dict):
                 rec for rec in result.get("recommendations", [])
                 if (
                     rec.action_plan in {"可操作", "重点关注"}
-                    or (rec.llm_score is not None and rec.llm_score >= 6)
                     or rec.score >= 56
                 )
             ]
@@ -916,7 +915,7 @@ async def _load_today_from_db() -> dict:
             result = await db.execute(
                 text("SELECT * FROM recommendations "
                      "WHERE date(created_at) = (SELECT date(created_at) FROM recommendations ORDER BY created_at DESC LIMIT 1) "
-                     "AND (action_plan IN ('可操作', '重点关注') OR COALESCE(llm_score, 0) >= 6 OR score >= 56) "
+                     "AND (action_plan IN ('可操作', '重点关注') OR score >= 56) "
                      "AND ("
                      "  COALESCE(recall_tags, '[]') LIKE '%hot_theme_core%' "
                      "  OR COALESCE(recall_tags, '[]') LIKE '%theme_leader%' "
diff --git a/backend/app/engine/scheduler.py b/backend/app/engine/scheduler.py
index 865f8fef..67bbc3dd 100644
--- a/backend/app/engine/scheduler.py
+++ b/backend/app/engine/scheduler.py
@@ -39,6 +39,26 @@ async def _run_scan(session_name: str):
         await log_error("scheduler", f"定时扫描失败 ({session_name}): {e}", detail=traceback.format_exc())
 
 
+async def _run_news_refresh(session_name: str = "scheduled"):
+    """后台采集新闻并更新主题催化分。"""
+    logger.info("=== 新闻催化刷新: %s ===", session_name)
+    try:
+        from app.news.pipeline import refresh_news_catalysts
+
+        result = await refresh_news_catalysts()
+        await broadcast_update({
+            "type": "news_catalysts_ready",
+            "session": session_name,
+            "inserted": result.get("inserted", 0),
+            "analyzed": result.get("analyzed", 0),
+            "timestamp": datetime.now().isoformat(),
+        })
+    except Exception as e:
+        logger.error(f"新闻催化刷新失败 ({session_name}): {e}")
+        from app.db.error_logger import log_error
+        await log_error("scheduler", f"新闻催化刷新失败 ({session_name}): {e}", detail=traceback.format_exc())
+
+
 async def _run_watchlist_analysis():
     """收盘后自动分析所有用户自选股。"""
     logger.info("=== 开始自选股定时分析 ===")
@@ -77,6 +97,22 @@ async def _run_strategy_iteration():
 def setup_scheduler():
     """配置所有定时任务（交易日时间）"""
 
+    news_schedule = [
+        ("news_pre_market", 8, 50, "pre_market"),
+        ("news_morning", 10, 5, "morning"),
+        ("news_noon", 12, 45, "noon"),
+        ("news_afternoon", 13, 55, "afternoon"),
+        ("news_post_market", 15, 40, "post_market"),
+    ]
+    for job_id, hour, minute, session_name in news_schedule:
+        scheduler.add_job(
+            _run_news_refresh,
+            CronTrigger(hour=hour, minute=minute, day_of_week="mon-fri"),
+            args=[session_name],
+            id=job_id,
+            replace_existing=True,
+        )
+
     # 盘前准备 09:00 - 计算前一日市场温度和板块数据
     scheduler.add_job(
         _run_scan, CronTrigger(hour=9, minute=0, day_of_week="mon-fri"),
diff --git a/backend/app/engine/screener.py b/backend/app/engine/screener.py
index 44aae676..ad5003d9 100644
--- a/backend/app/engine/screener.py
+++ b/backend/app/engine/screener.py
@@ -3,10 +3,10 @@
 三阶段管道：
   Step 1: 主线定位 — 把实时板块/快照板块归一成系统 MarketTheme
   Step 2: 主题内选股 — 从主线主题成分、领涨股和实时异动中召回候选
-  Step 3: 深度分析 — 资金顺势 + 供需 + 价格行为 + 趋势 + LLM
+  Step 3: 规则定价 — 催化 + 主题资金 + 个股资金 + 情绪角色 + 入场节奏
 
-评分公式：资金顺势 + 供需关系 + 价格行为 + 趋势。
-资金流和主题地位进入主评分，RSI/MACD 只作为节奏与风险参考。
+评分公式：市场热点/新闻催化 + 主线资金 + 个股资金 + 情绪地位 + 时机。
+技术指标只作为入场节奏与风控参考，不替代热点与资金主线。
 
 风险乘数：惩罚取最大而非叠加（防过度惩罚），奖励可叠加。
 
@@ -19,7 +19,6 @@
 
 import asyncio
 import logging
-import traceback
 
 import pandas as pd
 
@@ -144,29 +143,16 @@ async def run_screening(trade_date: str = None) -> dict:
         except Exception as e:
             logger.warning(f"注入实时价格失败，使用 Tushare 收盘价: {e}")
 
-    # ── Step 3: 规则边界 + LLM 两阶段裁决 ──
-    logger.info("=== Step 3: 规则边界 + LLM 两阶段裁决 ===")
+    # ── Step 3: 规则评分与交易计划 ──
+    logger.info("=== Step 3: 规则评分与交易计划 ===")
     recommendations = await _build_recommendations(
         candidates, market_temp, hot_sectors, market_temp_score, intraday, strategy_profile,
     )
 
-    if settings.deepseek_api_key:
-        recommendations = [
-            r for r in recommendations
-            if (
-                _is_main_theme_recommendation(r)
-                and (
-                    r.action_plan in {"可操作", "重点关注"}
-                    or (r.llm_score is not None and r.llm_score >= 6)
-                    or r.score >= max(strategy_profile.min_score - 4, 56)
-                )
-            )
-        ]
-    else:
-        recommendations = [
-            r for r in recommendations
-            if _is_main_theme_recommendation(r) and r.score >= strategy_profile.min_score
-        ]
+    recommendations = [
+        r for r in recommendations
+        if _is_main_theme_recommendation(r) and r.score >= strategy_profile.min_score
+    ]
 
     recommendations = _finalize_battle_plan(
         recommendations=recommendations,
@@ -388,7 +374,7 @@ async def _build_candidate_pool(
 ) -> list[dict]:
     """多路召回候选池。
 
-    目标是提高召回率，再交给 LLM 做资源分配与最终裁决。
+    目标是提高主线、资金、形态多路召回率，最终由规则评分统一排序。
     """
     merged: dict[str, dict] = {}
 
@@ -610,21 +596,16 @@ async def _build_recommendations(
     intraday: bool = False,
     strategy_profile=None,
 ) -> list[Recommendation]:
-    """Step 3: 规则边界建模 + LLM 两阶段裁决。"""
+    """Step 3: 规则边界建模、评分与交易计划生成。"""
     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_supply_demand,
-        analyze_volume_pattern,
         EntrySignal,
     )
     from app.analysis.signals import generate_signals
     from app.analysis.capital_flow import _score_valuation
-    from app.llm.batch_screener import (
-        analyze_candidates_individually,
-        prefilter_candidates_individually,
-    )
 
     # 名称和行业映射
     stock_basic = tushare_client.get_stock_basic()
@@ -636,7 +617,6 @@ async def _build_recommendations(
             industry_map[row["ts_code"]] = row.get("industry", "")
 
     recommendations = []
-    llm_candidates = []
     total = len(candidates)
     signal_counts = {"breakout": 0, "breakout_confirm": 0, "pullback": 0, "launch": 0, "reversal": 0, "none": 0}
     score_weights = strategy_profile.score_weights if strategy_profile else {
@@ -914,7 +894,6 @@ async def _build_recommendations(
             stock["entry_signal_type"] = effective_signal_name
             reasons = _generate_reasons(stock, entry_signal, tech_signal, df, intraday)
             risk_note = _generate_risk_note(market_temp, tech_signal, stock)
-            vol_pattern = analyze_volume_pattern(df)
             entry_timing = _generate_entry_timing(effective_signal_name, intraday)
             trade_plan = _build_trade_plan(
                 signal_type=effective_signal_name,
@@ -1001,55 +980,8 @@ async def _build_recommendations(
             )
             recommendations.append(rec)
 
-            # 收集 LLM 分析所需的候选摘要（不含 signal_type，让 LLM 独立判断）
-            llm_candidate = {
-                "ts_code": ts_code,
-                "name": name,
-                "sector": sector,
-                "quant_score": round(final_score, 1),
-                "position_score": round(position_score, 1),
-                "current_price": stock.get("price") or float(df.iloc[-1]["close"]),
-                "kline_summary": _summarize_for_llm(df, entry_signal, tech_signal),
-                "capital_flow_summary": (
-                    f"主力净流入{stock.get('main_net_inflow', 0):.0f}万, "
-                    f"占比{stock.get('inflow_ratio', 0):.1f}%"
-                ),
-                "recall_tags": stock.get("recall_tags", []),
-                "sector_stage": sector_stage,
-                "hot_theme_matched": bool(hot_theme_match),
-                "hot_theme_name": hot_theme_match.sector_name if hot_theme_match else "",
-                "hot_theme_aliases": hot_theme_match.theme_aliases if hot_theme_match else [],
-                "stock_role_hint": stock.get("stock_role_hint", "待判断"),
-                "entry_signal_type": effective_signal_name,
-                "entry_signal_score": round(entry_signal.get("signal_score", 0), 1),
-                "flow_momentum_score": round(flow_momentum_score, 1),
-                "signal_matches_profile": signal_matches_profile,
-                "risk_tags": risk_tags,
-                "focus_points": _build_focus_points(stock, entry_signal, tech_signal, vol_pattern, sector_stage),
-            }
-
-            if intraday:
-                try:
-                    from app.data.eastmoney_client import get_min_kline, analyze_intraday_volume_distribution
-                    min_df = await get_min_kline(ts_code, period="5", count=48)
-                    if not min_df.empty:
-                        vol_dist = analyze_intraday_volume_distribution(min_df)
-                        llm_candidate["intraday_volume"] = (
-                            f"上午量占比{vol_dist['morning_volume_ratio']}%, "
-                            f"下午{vol_dist['afternoon_volume_ratio']}%, "
-                            f"开盘30分{vol_dist['opening_strength']}%, "
-                            f"尾盘30分{vol_dist['closing_strength']}%, "
-                            f"趋势={vol_dist['volume_trend']}"
-                        )
-                        if vol_dist["key_periods"]:
-                            llm_candidate["intraday_volume"] += f", 放量时段: {'; '.join(vol_dist['key_periods'])}"
-                except Exception as e:
-                    logger.debug(f"分时量能数据获取失败 {ts_code}: {e}")
-
-            llm_candidates.append(llm_candidate)
-
         except Exception as e:
-            logger.debug(f"深度分析 {ts_code} 失败: {e}")
+            logger.debug(f"规则分析 {ts_code} 失败: {e}")
             continue
 
     logger.info(
@@ -1061,143 +993,6 @@ async def _build_recommendations(
     )
 
     recommendations.sort(key=lambda rec: rec.score, reverse=True)
-
-    if settings.deepseek_api_key and llm_candidates:
-        try:
-            market_summary = (
-                f"市场温度: {market_temp.temperature}/100, "
-                f"涨跌比: {market_temp.up_count}涨/{market_temp.down_count}跌, "
-                f"涨停: {market_temp.limit_up_count}家; "
-                f"今日主线主题: "
-                + " / ".join(
-                    f"{s.sector_name}[{' / '.join((s.theme_aliases or [])[:3])}]"
-                    f"({(s.realtime_pct_change if s.realtime_pct_change is not None else s.pct_change):+.2f}%)"
-                    for s in hot_sectors[:5]
-                )
-            )
-
-            llm_candidates.sort(key=lambda c: c["quant_score"], reverse=True)
-            prefilter_pool = llm_candidates[: settings.llm_prefilter_limit]
-            prefilter_results = await prefilter_candidates_individually(
-                prefilter_pool,
-                market_summary,
-                max_concurrent=settings.llm_prefilter_max_concurrent,
-            )
-
-            prioritized = []
-            for item in prefilter_pool:
-                pre = prefilter_results.get(item["ts_code"], {})
-                item["prefilter_decision"] = pre.get("decision", "watch")
-                item["prefilter_confidence"] = pre.get("confidence", 5)
-                item["prefilter_reason"] = pre.get("reason", "")
-                item["prefilter_focus_points"] = pre.get("focus_points", [])
-                if item["prefilter_decision"] == "priority":
-                    rank_bonus = 16
-                elif item["prefilter_decision"] == "watch":
-                    rank_bonus = 6
-                else:
-                    rank_bonus = -12
-                item["deep_rank"] = round(item["quant_score"] + rank_bonus + item["prefilter_confidence"] * 1.5, 1)
-                if item["prefilter_decision"] != "ignore":
-                    prioritized.append(item)
-
-            if not prioritized:
-                prioritized = prefilter_pool[: min(8, len(prefilter_pool))]
-
-            prioritized.sort(key=lambda c: c.get("deep_rank", c["quant_score"]), reverse=True)
-            llm_top = prioritized[: settings.llm_final_limit]
-            llm_results = await analyze_candidates_individually(llm_top, market_summary)
-
-            for rec in recommendations:
-                pre_item = next((item for item in prefilter_pool if item["ts_code"] == rec.ts_code), None)
-                if pre_item:
-                    rec.prefilter_decision = pre_item.get("prefilter_decision", "")
-                    rec.prefilter_reason = pre_item.get("prefilter_reason", "")
-                    rec.focus_points = pre_item.get("prefilter_focus_points", [])
-
-                llm_data = llm_results.get(rec.ts_code)
-                if llm_data:
-                    rec.llm_analysis = llm_data.get("analysis", "")
-                    rec.llm_score = float(llm_data.get("conviction", 0) or 0)
-
-                    verdict = llm_data.get("verdict", "watch")
-                    action_plan = llm_data.get("action_plan", "")
-                    conviction = float(llm_data.get("conviction", 6) or 6)
-                    ai_score = conviction * 10
-
-                    if verdict == "execute":
-                        rec.score = round(rec.score * 0.4 + ai_score * 0.6 + 4, 1)
-                    elif verdict == "watch":
-                        rec.score = round(rec.score * 0.5 + ai_score * 0.5 - 2, 1)
-                    else:  # skip
-                        rec.score = round(rec.score * 0.45 + ai_score * 0.35 - 18, 1)
-
-                    if verdict == "skip":
-                        rec.signal = "HOLD"
-                        rec.action_plan = "观察"
-                        rec.lifecycle_status = "candidate"
-                        if not rec.risk_note:
-                            rec.risk_note = llm_data.get("risk_flag", "") or rec.risk_note
-                    else:
-                        if action_plan in {"可操作", "重点关注", "观察"}:
-                            rec.action_plan = action_plan
-                        elif verdict == "execute":
-                            rec.action_plan = "可操作"
-                        else:
-                            rec.action_plan = "重点关注"
-
-                        rec.signal = "BUY" if verdict == "execute" else "HOLD"
-                        if rec.action_plan == "可操作":
-                            rec.lifecycle_status = "actionable"
-                        elif rec.action_plan == "重点关注":
-                            rec.lifecycle_status = "candidate"
-
-                    if llm_data.get("timing"):
-                        rec.entry_timing = llm_data["timing"]
-
-                    if llm_data.get("trigger_condition"):
-                        rec.trigger_condition = llm_data["trigger_condition"]
-                    if llm_data.get("invalidation_condition"):
-                        rec.invalidation_condition = llm_data["invalidation_condition"]
-                    if llm_data.get("position_pct") is not None:
-                        rec.suggested_position_pct = float(llm_data["position_pct"] or 0)
-                    if llm_data.get("risk_flag"):
-                        rec.risk_note = llm_data["risk_flag"]
-
-                    rec.level = _score_to_level(rec.score)
-                    _apply_llm_trace(
-                        rec,
-                        verdict=verdict,
-                        action_plan=rec.action_plan,
-                        conviction=conviction,
-                        reason=llm_data.get("analysis", "") or llm_data.get("risk_flag", ""),
-                    )
-
-                    # 用 LLM 给出的价格替代结构化规则价格
-                    if llm_data.get("entry_price"):
-                        rec.entry_price = llm_data["entry_price"]
-                    if llm_data.get("target_price"):
-                        rec.target_price = llm_data["target_price"]
-                    if llm_data.get("stop_loss"):
-                        rec.stop_loss = llm_data["stop_loss"]
-
-            recommendations = [
-                rec for rec in recommendations
-                if not (
-                    rec.llm_score is not None
-                    and rec.llm_score <= 4
-                    and rec.action_plan == "观察"
-                    and rec.score < max(strategy_profile.min_score - 6, 54)
-                )
-            ]
-            recommendations.sort(key=lambda r: r.score, reverse=True)
-            recommendations = recommendations[:settings.top_stock_count]
-            logger.info(f"LLM 两阶段分析完成, 综合评分后保留 {len(recommendations)} 只")
-        except Exception as e:
-            logger.error(f"LLM 两阶段分析失败, 仅使用规则边界: {e}")
-            from app.db.error_logger import log_error
-            await log_error("screener", f"LLM 两阶段分析失败, 仅使用规则边界: {e}", detail=traceback.format_exc())
-
     return recommendations
 
 
@@ -2224,186 +2019,3 @@ def _build_trace_evidence(
         evidence.append("符合今日策略偏好的入场类型")
     return evidence[:5]
 
-
-def _apply_llm_trace(
-    rec: Recommendation,
-    verdict: str,
-    action_plan: str,
-    conviction: float,
-    reason: str,
-) -> None:
-    trace = dict(rec.decision_trace or {})
-    trace["llm_adjustment"] = {
-        "verdict": verdict,
-        "action_plan": action_plan,
-        "conviction": round(conviction, 1),
-        "reason": str(reason or "")[:180],
-    }
-    trace["action_plan"] = action_plan
-    if verdict == "execute":
-        trace["headline"] = f"AI确认可执行: {trace.get('headline', rec.name)}"
-    elif verdict == "skip":
-        trace["headline"] = f"AI降级观察: {trace.get('headline', rec.name)}"
-    rec.decision_trace = trace
-
-
-def _build_focus_points(
-    stock: dict,
-    entry_signal: dict,
-    tech: TechnicalSignal | None,
-    vol_pattern: dict,
-    sector_stage: str,
-) -> list[str]:
-    points: list[str] = []
-    signal_type = entry_signal.get("signal_type")
-    if signal_type and getattr(signal_type, "value", "none") != "none":
-        points.append(f"确认{signal_type.value}信号是否延续")
-    elif stock.get("entry_signal_type") == "flow_momentum":
-        points.append("确认主力流入和板块前排强度是否延续")
-    if stock.get("main_net_inflow", 0) > 0:
-        points.append("观察主力流入是否继续放大")
-    if vol_pattern.get("volume_trend"):
-        points.append(f"量能状态: {vol_pattern['volume_trend']}")
-    if tech and tech.support_price:
-        points.append(f"关键支撑 {tech.support_price}")
-    if sector_stage in ("late", "end"):
-        points.append("板块已偏后段，注意是否还有前排承接")
-    return points[:4]
-
-
-def _summarize_for_llm(df, entry_signal: dict, tech_signal: TechnicalSignal | None) -> str:
-    """生成 K 线分析结论供 LLM 判断（输出结论而非原始数据）"""
-    import pandas as pd
-
-    last = df.iloc[-1]
-    parts = []
-
-    # ── 趋势结论 ──
-    ma_fields = ["ma5", "ma10", "ma20", "ma60"]
-    ma_vals = {m: last.get(m) for m in ma_fields}
-
-    trend_desc = "趋势不明"
-    all_ma_valid = all(ma_vals.get(m) is not None and not pd.isna(ma_vals[m]) for m in ma_fields)
-    if all_ma_valid:
-        if ma_vals["ma5"] > ma_vals["ma10"] > ma_vals["ma20"] > ma_vals["ma60"]:
-            trend_desc = "强势多头排列(MA5>MA10>MA20>MA60)"
-        elif ma_vals["ma5"] > ma_vals["ma10"] > ma_vals["ma20"]:
-            trend_desc = "中短期多头(MA5>MA10>MA20)"
-        elif ma_vals["ma5"] > ma_vals["ma20"]:
-            trend_desc = "偏多(MA5在MA20上方)"
-        elif ma_vals["ma5"] < ma_vals["ma10"] < ma_vals["ma20"]:
-            trend_desc = "空头排列，趋势偏弱"
-        else:
-            trend_desc = "均线交织，趋势震荡"
-
-    # MA20 方向
-    if len(df) >= 5 and not pd.isna(last.get("ma20")) and not pd.isna(df.iloc[-5].get("ma20")):
-        ma20_now = last["ma20"]
-        ma20_5d = df.iloc[-5]["ma20"]
-        if ma20_5d > 0:
-            ma20_pct = (ma20_now - ma20_5d) / ma20_5d * 100
-            if ma20_pct > 2:
-                trend_desc += "，MA20快速上扬"
-            elif ma20_pct > 0:
-                trend_desc += "，MA20缓慢上行"
-            else:
-                trend_desc += "，MA20走平或下行"
-    parts.append(trend_desc)
-
-    # ── 量价结论 ──
-    if len(df) >= 10:
-        recent = df.tail(10)
-        up_days = recent[recent["pct_chg"] > 0]
-        down_days = recent[recent["pct_chg"] <= 0]
-        vol_conclusion = ""
-        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 > 1.5:
-                    vol_conclusion = f"量价健康(上涨均量/下跌均量={ratio:.1f}，需求主导)"
-                elif ratio > 1.0:
-                    vol_conclusion = f"量价尚可(量比={ratio:.1f}，需求略强)"
-                else:
-                    vol_conclusion = f"量价偏弱(量比={ratio:.1f}，供给主导)"
-        if not vol_conclusion:
-            vol_conclusion = "量价数据不足"
-
-        # 最近5日量能变化
-        recent_5 = df.tail(5)
-        vol_ma5 = recent_5["vol"].mean()
-        vol_ma10 = df.tail(10)["vol"].mean()
-        if vol_ma10 > 0:
-            vol_ratio = vol_ma5 / vol_ma10
-            if vol_ratio > 1.5:
-                vol_conclusion += "，近5日明显放量"
-            elif vol_ratio < 0.7:
-                vol_conclusion += "，近5日缩量"
-        parts.append(vol_conclusion)
-
-    # ── MACD 结论（节奏参考） ──
-    dif = last.get("dif", 0) or 0
-    dea = last.get("dea", 0) or 0
-    macd_desc = ""
-    if len(df) >= 3:
-        prev_dif = df.iloc[-2].get("dif", 0) or 0
-        prev_dea = df.iloc[-2].get("dea", 0) or 0
-        if dif > dea and prev_dif <= prev_dea:
-            macd_desc = "MACD刚金叉"
-        elif dif > dea:
-            macd_desc = "MACD金叉运行中"
-        elif dif < dea and prev_dif >= prev_dea:
-            macd_desc = "MACD刚死叉"
-        elif dif < dea:
-            macd_desc = "MACD死叉运行中"
-
-        if dif > 0:
-            macd_desc += "，零轴上方(偏多)"
-        else:
-            macd_desc += "，零轴下方(偏空)"
-    parts.append((macd_desc or "MACD数据不足") + "；仅作节奏参考")
-
-    # ── RSI 结论（风险提示，不做买卖裁判） ──
-    rsi = last.get("rsi14", 50)
-    if not pd.isna(rsi):
-        if rsi > 80:
-            parts.append(f"RSI14={rsi:.0f}，偏热，提示追高风险但不单独否决资金顺势")
-        elif rsi > 70:
-            parts.append(f"RSI14={rsi:.0f}，偏高，注意追高风险")
-        elif rsi >= 40:
-            parts.append(f"RSI14={rsi:.0f}，节奏中性")
-        else:
-            parts.append(f"RSI14={rsi:.0f}，偏低，提示弱势或反弹弹性，不单独构成买点")
-
-    # ── 价格位置结论 ──
-    if tech_signal:
-        pos_parts = []
-        if tech_signal.rally_pct_5d > 15:
-            pos_parts.append(f"5日已涨{tech_signal.rally_pct_5d}%，追高风险大")
-        elif tech_signal.rally_pct_5d > 8:
-            pos_parts.append(f"5日涨{tech_signal.rally_pct_5d}%，短期有一定涨幅")
-        elif tech_signal.rally_pct_5d > 0:
-            pos_parts.append(f"5日涨{tech_signal.rally_pct_5d}%，涨幅温和")
-        else:
-            pos_parts.append(f"5日跌{abs(tech_signal.rally_pct_5d)}%，回调中")
-
-        if tech_signal.distance_from_high >= 0:
-            pos_parts.append("处于60日新高附近")
-        elif tech_signal.distance_from_high > -5:
-            pos_parts.append(f"距60日高点{abs(tech_signal.distance_from_high):.1f}%")
-        else:
-            pos_parts.append(f"距60日高点{abs(tech_signal.distance_from_high):.1f}%，位置较低")
-
-        parts.append("位置: " + "，".join(pos_parts))
-
-    # ── 近5日价格走势简述 ──
-    if len(df) >= 5:
-        recent_5 = df.tail(5)
-        closes = recent_5["close"].tolist()
-        first_c = closes[0]
-        last_c = closes[-1]
-        pct_5d = (last_c - first_c) / first_c * 100
-        parts.append(f"当前价: {last_c:.2f}，5日{'涨' if pct_5d >= 0 else '跌'}{abs(pct_5d):.1f}%")
-
-    return "\n".join(parts)
diff --git a/backend/app/llm/batch_screener.py b/backend/app/llm/batch_screener.py
index 4f4ec546..ba5e8b4d 100644
--- a/backend/app/llm/batch_screener.py
+++ b/backend/app/llm/batch_screener.py
@@ -1,4 +1,9 @@
-"""LLM 候选预筛 + 逐股深度分析
+"""历史 LLM 候选预筛 + 逐股深度分析工具。
+
+弃用说明：
+生产推荐链路禁止调用本模块。A 股批量选股必须由热点/催化、资金流、
+情绪地位和规则化交易计划决定，LLM 只能做用户主动触发的解释、会诊、
+离线复盘或新闻催化归因。
 
 先做轻量预筛，控制深度裁决成本；
 再对重点股票单独调用 LLM 做深度分析。
diff --git a/backend/app/llm/prompts.py b/backend/app/llm/prompts.py
index 8228ade5..58e03eed 100644
--- a/backend/app/llm/prompts.py
+++ b/backend/app/llm/prompts.py
@@ -42,15 +42,17 @@ CHAT_SYSTEM_PROMPT = """\
 3. 查询当前用户的自选股池与最新建议
 4. 查询个股K线、技术面、资金流向数据
 5. 搜索股票代码，并把结果放回当前交易语境中分析
-6. 对单只股票生成系统化会诊，输出结论、触发条件、失效条件、仓位边界和风险清单
+6. 在用户明确要求时，对单只股票生成系统化会诊，输出结论、触发条件、失效条件、仓位边界和风险清单
 
 重要提醒：
 - 回答用户关于"今天市场怎么样"之类的问题时，必须调用 get_realtime_indices 获取实时指数数据
 - 回答用户关于"今天该怎么做"、"当前推荐怎么看"、"自选股该怎么处理"这类问题时，优先调用 get_strategy_board、get_latest_recommendations、get_user_watchlist_snapshot
-- 回答用户关于某只股票能不能看、是否该买、持仓怎么处理、为什么涨跌、是否要复盘时，必须先 search_stock（如果用户没给标准 ts_code），再调用 diagnose_stock；必要时补充 get_stock_capital_flow、get_stock_technical_signal
+- 回答用户关于某只股票时，先 search_stock（如果用户没给标准 ts_code），再用 get_latest_recommendations、get_user_watchlist_snapshot、get_stock_capital_flow、get_stock_technical_signal 等工具读取现有证据
+- 只有当用户明确要求"诊断"、"会诊"、"深度分析"、"生成报告"或正在使用诊断页面时，才调用 diagnose_stock；不要在普通聊天、列表解释、推荐解读中自动触发个股会诊
 - 盘中时段（9:30-15:00）必须使用实时数据，盘后时段使用当日收盘或最近一次系统生成的数据
 - 不要脱离系统上下文泛泛而谈，必须先调用工具获取最新结果再回答
 - 个股分析必须优先看资金流向、主线板块、量价关系、价格行为和位置边界；RSI/MACD/KDJ 只做最后的节奏与风控备注，不能因超买超卖本身直接否决或买入
+- LLM 只负责解释、归纳和生成用户主动请求的诊断文本，不参与批量选股、排序、行动计划、止盈止损或策略换挡
 
 回答要求：
 1. 使用工具获取最新数据后再回答，不要凭空编造数据
diff --git a/backend/app/llm/strategy_selector.py b/backend/app/llm/strategy_selector.py
index ebbaf3f6..848fc18d 100644
--- a/backend/app/llm/strategy_selector.py
+++ b/backend/app/llm/strategy_selector.py
@@ -1,7 +1,8 @@
 """动态策略选择器
 
 在固定筛选引擎前增加一层“先选打法，再选股票”的策略决策。
-规则负责稳定分类，LLM 负责补充语义判断和操作建议。
+生产筛选只使用规则和策略配置，保证同一份行情输入得到稳定输出。
+LLM 只能用于离线复盘、配置建议或解释，不参与盘中策略换挡。
 """
 
 import json
@@ -125,14 +126,7 @@ async def select_strategy_profile(
     from app.llm.strategy_config import load_active_strategy_profile
 
     profile = _select_rule_profile(market_temp, hot_sectors, intraday)
-    profile = await load_active_strategy_profile(profile)
-
-    if settings.deepseek_api_key:
-        llm_profile = await _select_llm_profile(market_temp, hot_sectors, intraday, profile)
-        if llm_profile:
-            profile = await load_active_strategy_profile(llm_profile)
-
-    return profile
+    return await load_active_strategy_profile(profile)
 
 
 def _select_rule_profile(
diff --git a/backend/app/news/__init__.py b/backend/app/news/__init__.py
new file mode 100644
index 00000000..97d2d0d1
--- /dev/null
+++ b/backend/app/news/__init__.py
@@ -0,0 +1,5 @@
+"""新闻源采集层。
+
+该层只负责把外部新闻落到本地数据库。题材归因由 catalyst 层处理，
+页面和普通 API 不应直接触发外部新闻抓取。
+"""
diff --git a/backend/app/news/collector.py b/backend/app/news/collector.py
new file mode 100644
index 00000000..1c2f80a9
--- /dev/null
+++ b/backend/app/news/collector.py
@@ -0,0 +1,189 @@
+"""多源新闻采集器。"""
+
+from __future__ import annotations
+
+import email.utils
+import html
+import logging
+import re
+import xml.etree.ElementTree as ET
+from datetime import datetime, timedelta
+
+import httpx
+
+from app.config import settings
+from app.data.tushare_client import tushare_client
+from app.news.models import NewsItem
+
+logger = logging.getLogger(__name__)
+_tushare_source_cursor = 0
+
+RSS_HEADERS = {
+    "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36",
+}
+
+
+async def collect_news_sources(
+    lookback_hours: int | None = None,
+    limit_per_source: int | None = None,
+) -> list[NewsItem]:
+    """从配置的数据源采集新闻。失败源隔离，不影响其他源。"""
+    lookback_hours = lookback_hours or settings.news_fetch_lookback_hours
+    limit_per_source = limit_per_source or settings.news_fetch_limit_per_source
+    items: list[NewsItem] = []
+
+    for source in _select_tushare_sources_for_run():
+        try:
+            items.extend(await _collect_tushare_news(source, lookback_hours, limit_per_source))
+        except Exception as e:
+            logger.warning("Tushare 新闻源采集失败 source=%s error=%s", source, e)
+
+    rss_sources = _parse_rss_sources(settings.news_rss_sources)
+    if rss_sources:
+        async with httpx.AsyncClient(headers=RSS_HEADERS, timeout=10.0, follow_redirects=True) as client:
+            for name, url in rss_sources:
+                try:
+                    items.extend(await _collect_rss(client, name, url, lookback_hours, limit_per_source))
+                except Exception as e:
+                    logger.warning("RSS 新闻源采集失败 source=%s url=%s error=%s", name, url, e)
+
+    return _dedup_in_memory(items)
+
+
+async def _collect_tushare_news(source: str, lookback_hours: int, limit: int) -> list[NewsItem]:
+    df = tushare_client.get_news(
+        source=source,
+        start_time=datetime.now() - timedelta(hours=lookback_hours),
+        end_time=datetime.now(),
+        limit=limit,
+    )
+    if df.empty:
+        return []
+
+    items: list[NewsItem] = []
+    for _, row in df.iterrows():
+        title = _clean_text(row.get("title", ""))
+        if len(title) < settings.news_min_title_length:
+            continue
+        content = _clean_text(row.get("content", ""))
+        items.append(NewsItem(
+            title=title,
+            content=content,
+            summary=_clean_text(row.get("summary", "")),
+            source=f"tushare:{source}",
+            url=str(row.get("url", "") or ""),
+            published_at=_parse_datetime(row.get("datetime") or row.get("time") or row.get("publish_time")),
+        ))
+    return items[:limit]
+
+
+async def _collect_rss(
+    client: httpx.AsyncClient,
+    source: str,
+    url: str,
+    lookback_hours: int,
+    limit: int,
+) -> list[NewsItem]:
+    resp = await client.get(url)
+    resp.raise_for_status()
+    root = ET.fromstring(resp.content)
+    cutoff = datetime.now() - timedelta(hours=lookback_hours)
+    items: list[NewsItem] = []
+
+    for item in root.findall(".//item")[: limit * 2]:
+        title = _clean_text(_xml_text(item, "title"))
+        if len(title) < settings.news_min_title_length:
+            continue
+        published_at = _parse_datetime(_xml_text(item, "pubDate"))
+        if published_at and published_at < cutoff:
+            continue
+        summary = _clean_text(_xml_text(item, "description"))
+        items.append(NewsItem(
+            title=title,
+            content=summary,
+            summary=summary[:240],
+            source=f"rss:{source}",
+            url=_clean_text(_xml_text(item, "link")),
+            published_at=published_at,
+        ))
+        if len(items) >= limit:
+            break
+
+    return items
+
+
+def _split_csv(value: str) -> list[str]:
+    return [item.strip() for item in (value or "").split(",") if item.strip()]
+
+
+def _parse_rss_sources(value: str) -> list[tuple[str, str]]:
+    result: list[tuple[str, str]] = []
+    for chunk in _split_csv(value):
+        if "|" not in chunk:
+            continue
+        name, url = chunk.split("|", 1)
+        name = name.strip()
+        url = url.strip()
+        if name and url:
+            result.append((name, url))
+    return result
+
+
+def _select_tushare_sources_for_run() -> list[str]:
+    """Tushare news 免费/低权限账号通常限制 1 次/分钟，每轮只取少量源。"""
+    global _tushare_source_cursor
+
+    sources = _split_csv(settings.news_tushare_sources)
+    if not sources:
+        return []
+
+    limit = max(1, min(int(settings.news_tushare_sources_per_run or 1), len(sources)))
+    selected: list[str] = []
+    for offset in range(limit):
+        selected.append(sources[(_tushare_source_cursor + offset) % len(sources)])
+    _tushare_source_cursor = (_tushare_source_cursor + limit) % len(sources)
+    return selected
+
+
+def _xml_text(item: ET.Element, tag: str) -> str:
+    node = item.find(tag)
+    return node.text if node is not None and node.text else ""
+
+
+def _clean_text(value) -> str:
+    text = html.unescape(str(value or ""))
+    text = re.sub(r"<[^>]+>", " ", text)
+    text = re.sub(r"\s+", " ", text)
+    return text.strip()
+
+
+def _parse_datetime(value) -> datetime | None:
+    if value is None:
+        return None
+    if isinstance(value, datetime):
+        return value.replace(tzinfo=None)
+    text = str(value).strip()
+    if not text:
+        return None
+    for fmt in ("%Y-%m-%d %H:%M:%S", "%Y-%m-%d %H:%M", "%Y%m%d%H%M%S", "%Y%m%d"):
+        try:
+            return datetime.strptime(text[: len(fmt)], fmt)
+        except Exception:
+            pass
+    try:
+        parsed = email.utils.parsedate_to_datetime(text)
+        return parsed.replace(tzinfo=None)
+    except Exception:
+        return None
+
+
+def _dedup_in_memory(items: list[NewsItem]) -> list[NewsItem]:
+    seen: set[str] = set()
+    result: list[NewsItem] = []
+    for item in items:
+        key = re.sub(r"\W+", "", item.title.lower())[:80]
+        if not key or key in seen:
+            continue
+        seen.add(key)
+        result.append(item)
+    return result
diff --git a/backend/app/news/models.py b/backend/app/news/models.py
new file mode 100644
index 00000000..ded5d7fa
--- /dev/null
+++ b/backend/app/news/models.py
@@ -0,0 +1,13 @@
+"""新闻采集领域模型。"""
+
+from datetime import datetime
+from pydantic import BaseModel
+
+
+class NewsItem(BaseModel):
+    title: str
+    content: str = ""
+    summary: str = ""
+    source: str = ""
+    url: str = ""
+    published_at: datetime | None = None
diff --git a/backend/app/news/pipeline.py b/backend/app/news/pipeline.py
new file mode 100644
index 00000000..ef1b3982
--- /dev/null
+++ b/backend/app/news/pipeline.py
@@ -0,0 +1,50 @@
+"""新闻采集与催化归因流水线。"""
+
+from __future__ import annotations
+
+import logging
+
+from app.config import settings
+from app.catalyst.service import analyze_pending_news, ingest_news_items
+from app.news.collector import collect_news_sources
+
+logger = logging.getLogger(__name__)
+
+
+async def refresh_news_catalysts(
+    lookback_hours: int | None = None,
+    limit_per_source: int | None = None,
+    analyze_limit: int | None = None,
+    use_llm: bool = True,
+) -> dict:
+    if not settings.news_collection_enabled:
+        return {
+            "enabled": False,
+            "fetched": 0,
+            "inserted": 0,
+            "duplicates": 0,
+            "analyzed": 0,
+            "skipped": 0,
+            "failed": 0,
+        }
+
+    items = await collect_news_sources(
+        lookback_hours=lookback_hours or settings.news_fetch_lookback_hours,
+        limit_per_source=limit_per_source or settings.news_fetch_limit_per_source,
+    )
+    ingest_result = await ingest_news_items(items)
+    remaining_result = await analyze_pending_news(
+        limit=analyze_limit or settings.news_analyze_limit_per_run,
+        use_llm=use_llm,
+    )
+    result = {
+        "enabled": True,
+        "fetched": len(items),
+        "inserted": ingest_result.get("inserted", 0),
+        "duplicates": ingest_result.get("duplicates", 0),
+        "analyzed": ingest_result.get("analyzed", 0) + remaining_result.get("analyzed", 0),
+        "skipped": remaining_result.get("skipped", 0),
+        "failed": ingest_result.get("failed", 0) + remaining_result.get("failed", 0),
+    }
+    logger.info("新闻催化刷新完成: %s", result)
+    return result
diff --git a/frontend/src/app/(auth)/dashboard/page.tsx b/frontend/src/app/(auth)/dashboard/page.tsx
index 2b1e97a9..b383e523 100644
--- a/frontend/src/app/(auth)/dashboard/page.tsx
+++ b/frontend/src/app/(auth)/dashboard/page.tsx
@@ -606,7 +606,7 @@ function FocusStockCard({ rec }: { rec: RecommendationData }) {
         </div>
 
         <div className="mt-1.5 text-xs leading-5 text-text-secondary line-clamp-2">
-          {rec.decision_trace?.headline ?? rec.trigger_condition ?? rec.entry_timing ?? rec.prefilter_reason ?? rec.reasons?.[0] ?? "等待新的触发条件。"}
+          {rec.decision_trace?.headline ?? rec.trigger_condition ?? rec.entry_timing ?? rec.reasons?.[0] ?? "等待新的触发条件。"}
         </div>
 
         {(rec.invalidation_condition || rec.risk_note) ? (
diff --git a/frontend/src/app/(auth)/sentiment/page.tsx b/frontend/src/app/(auth)/sentiment/page.tsx
new file mode 100644
index 00000000..a9c1ea81
--- /dev/null
+++ b/frontend/src/app/(auth)/sentiment/page.tsx
@@ -0,0 +1,429 @@
+"use client";
+
+import { useCallback, useEffect, useMemo, useState } from "react";
+import { ErrorBoundary } from "@/components/error-boundary";
+import { fetchAPI } from "@/lib/api";
+import type { CatalystEvent, CatalystNewsItem, ThemeCatalystScore } from "@/lib/api";
+import { useWebSocket } from "@/hooks/use-websocket";
+
+type CatalystTone = "hot" | "warm" | "quiet";
+
+export default function SentimentPage() {
+  const [news, setNews] = useState<CatalystNewsItem[]>([]);
+  const [events, setEvents] = useState<CatalystEvent[]>([]);
+  const [themes, setThemes] = useState<ThemeCatalystScore[]>([]);
+  const [loading, setLoading] = useState(true);
+  const [statusFilter, setStatusFilter] = useState("all");
+
+  const loadData = useCallback(async () => {
+    try {
+      const [newsData, eventData, themeData] = await Promise.all([
+        fetchAPI<CatalystNewsItem[]>("/api/catalysts/news?limit=80&hours=72").catch(() => []),
+        fetchAPI<CatalystEvent[]>("/api/catalysts/recent?limit=60&hours=72").catch(() => []),
+        fetchAPI<ThemeCatalystScore[]>("/api/catalysts/theme-scores?limit=20&hours=72").catch(() => []),
+      ]);
+      setNews(newsData);
+      setEvents(eventData);
+      setThemes(themeData);
+    } finally {
+      setLoading(false);
+    }
+  }, []);
+
+  useEffect(() => {
+    loadData();
+  }, [loadData]);
+
+  useWebSocket(
+    useCallback((message: { type: string }) => {
+      if (message.type === "news_catalysts_ready" || message.type === "sector_scan_ready" || message.type === "scan_complete") {
+        loadData();
+      }
+    }, [loadData])
+  );
+
+  const analyzedCount = news.filter((item) => item.status === "analyzed").length;
+  const pendingCount = news.filter((item) => item.status === "pending").length;
+  const failedCount = news.filter((item) => item.status === "failed").length;
+  const topTheme = themes[0];
+  const headline = buildSentimentHeadline(themes, events);
+
+  const filteredNews = useMemo(() => {
+    if (statusFilter === "all") return news;
+    return news.filter((item) => item.status === statusFilter);
+  }, [news, statusFilter]);
+
+  const eventsById = useMemo(() => {
+    const map = new Map<number, CatalystEvent>();
+    events.forEach((event) => map.set(event.id, event));
+    return map;
+  }, [events]);
+
+  return (
+    <ErrorBoundary>
+      <div className="max-w-7xl mx-auto px-4 md:px-8 pt-6 pb-20 md:pb-10 space-y-5">
+        <div className="animate-fade-in-up">
+          <h1 className="text-lg font-bold tracking-tight">舆情雷达</h1>
+        </div>
+
+        <div className="grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_340px] gap-4 animate-fade-in-up">
+          <section className="glass-card-static p-4 md:p-5">
+            <div className="flex flex-col gap-4 md:flex-row md:items-start md:justify-between">
+              <div className="min-w-0">
+                <div className="text-[10px] uppercase tracking-[0.22em] text-amber-400 font-semibold">市场情绪线索</div>
+                <h2 className="mt-2 text-xl font-bold tracking-tight text-text-primary">{headline.title}</h2>
+                <p className="mt-2 max-w-3xl text-sm leading-6 text-text-secondary">{headline.detail}</p>
+              </div>
+              <div className="grid grid-cols-2 gap-2 md:w-[280px] shrink-0">
+                <RadarMetric label="已分析" value={analyzedCount} tone="hot" />
+                <RadarMetric label="待处理" value={pendingCount} tone="warm" />
+                <RadarMetric label="强主题" value={themes.filter((item) => item.catalyst_score >= 70).length} tone="hot" />
+                <RadarMetric label="异常" value={failedCount} tone={failedCount ? "warm" : "quiet"} />
+              </div>
+            </div>
+
+            <div className="mt-4 grid grid-cols-1 md:grid-cols-3 gap-3">
+              <RadarDecision title="当前重点" value={topTheme?.theme_name ?? "暂无"} detail={topTheme ? `催化分 ${topTheme.catalyst_score.toFixed(0)}，${topTheme.catalyst_count} 条线索` : "等待后台新闻归因。"} tone="hot" />
+              <RadarDecision title="观察方式" value={headline.action} detail="只作为主题热度和催化背景，不直接生成买卖结论。" tone="warm" />
+              <RadarDecision title="失效信号" value={headline.risk} detail="当新闻无法映射到资金主线时，降低权重。" tone="quiet" />
+            </div>
+          </section>
+
+          <section className="glass-card-static p-4">
+            <div className="flex items-center justify-between gap-3">
+              <h2 className="text-sm font-semibold text-text-primary">主题催化排行</h2>
+              <span className="text-[10px] text-text-muted">72小时</span>
+            </div>
+            <div className="mt-3 space-y-2.5">
+              {themes.length ? themes.slice(0, 8).map((theme, index) => (
+                <ThemePulseRow key={theme.theme_id} theme={theme} rank={index + 1} />
+              )) : (
+                <EmptyLine text={loading ? "加载舆情主题..." : "暂无主题催化"} />
+              )}
+            </div>
+          </section>
+        </div>
+
+        <div className="grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_380px] gap-4 animate-fade-in-up">
+          <section className="glass-card-static p-4 md:p-5">
+            <div className="flex flex-wrap items-center justify-between gap-3">
+              <div>
+                <h2 className="text-sm font-semibold text-text-primary">舆情流</h2>
+                <p className="mt-1 text-xs text-text-muted">后台抓取的新闻、政策和公告线索。</p>
+              </div>
+              <div className="flex flex-wrap gap-2">
+                {[
+                  { key: "all", label: "全部", count: news.length },
+                  { key: "analyzed", label: "已归因", count: analyzedCount },
+                  { key: "pending", label: "待处理", count: pendingCount },
+                  { key: "failed", label: "异常", count: failedCount },
+                ].map((tab) => (
+                  <button
+                    key={tab.key}
+                    onClick={() => setStatusFilter(tab.key)}
+                    className={`rounded-lg border px-3 py-1.5 text-xs font-medium transition-all ${
+                      statusFilter === tab.key
+                        ? "border-amber-500/20 bg-amber-500/10 text-amber-300"
+                        : "border-border-subtle bg-surface-1 text-text-muted hover:text-text-primary"
+                    }`}
+                  >
+                    {tab.label} {tab.count}
+                  </button>
+                ))}
+              </div>
+            </div>
+
+            <div className="mt-4 space-y-3">
+              {filteredNews.length ? filteredNews.map((item) => (
+                <NewsRow key={item.id} item={item} event={item.catalyst_id ? eventsById.get(item.catalyst_id) : undefined} />
+              )) : (
+                <div className="rounded-2xl border border-border-subtle bg-surface-1/70 p-8 text-center text-sm text-text-muted">
+                  {loading ? "加载舆情流..." : "暂无符合条件的舆情。"}
+                </div>
+              )}
+            </div>
+          </section>
+
+          <aside className="space-y-4">
+            <section className="glass-card-static p-4">
+              <h2 className="text-sm font-semibold text-text-primary">AI 归因摘要</h2>
+              <div className="mt-3 space-y-3">
+                {events.length ? events.slice(0, 6).map((event) => (
+                  <CatalystInsight key={event.id} event={event} />
+                )) : (
+                  <EmptyLine text={loading ? "加载归因结果..." : "暂无 AI 归因结果"} />
+                )}
+              </div>
+            </section>
+
+            <section className="glass-card-static p-4">
+              <h2 className="text-sm font-semibold text-text-primary">使用边界</h2>
+              <div className="mt-3 space-y-2 text-xs leading-6 text-text-secondary">
+                <BoundaryLine text="舆情只解释催化方向，不直接给出买卖。" />
+                <BoundaryLine text="主题催化需要和资金流、前排强度共同确认。" />
+                <BoundaryLine text="页面只读取本地数据库，不触发外部抓取。" />
+              </div>
+            </section>
+          </aside>
+        </div>
+      </div>
+    </ErrorBoundary>
+  );
+}
+
+function buildSentimentHeadline(themes: ThemeCatalystScore[], events: CatalystEvent[]) {
+  const top = themes[0];
+  const strongCount = themes.filter((item) => item.catalyst_score >= 70).length;
+  const policyCount = events.filter((item) => item.catalyst_type === "policy").length;
+
+  if (!top) {
+    return {
+      title: "舆情等待后台归因",
+      detail: "新闻采集和催化分析会在后台任务完成后更新。",
+      action: "等待线索",
+      risk: "无数据不判断",
+    };
+  }
+
+  if (top.catalyst_score >= 75) {
+    return {
+      title: `${top.theme_name} 舆情升温`,
+      detail: `过去 72 小时 ${top.catalyst_count} 条催化线索，${strongCount} 个主题达到强催化阈值。${policyCount ? `其中政策类线索 ${policyCount} 条。` : ""}`,
+      action: "优先核对资金回流",
+      risk: "热度兑现",
+    };
+  }
+
+  return {
+    title: `${top.theme_name} 有线索但未形成强共振`,
+    detail: `当前催化分 ${top.catalyst_score.toFixed(0)}，更适合观察是否被资金和板块前排确认。`,
+    action: "跟踪扩散",
+    risk: "证据不足",
+  };
+}
+
+function RadarMetric({ label, value, tone }: { label: string; value: number; tone: CatalystTone }) {
+  const toneClass = getToneClass(tone);
+  return (
+    <div className={`rounded-xl border px-3 py-2 ${toneClass.box}`}>
+      <div className="text-[10px] text-text-muted">{label}</div>
+      <div className={`mt-1 font-mono text-xl font-bold tabular-nums ${toneClass.text}`}>{value}</div>
+    </div>
+  );
+}
+
+function RadarDecision({ title, value, detail, tone }: { title: string; value: string; detail: string; tone: CatalystTone }) {
+  const toneClass = getToneClass(tone);
+  return (
+    <div className={`rounded-2xl border p-3 ${toneClass.box}`}>
+      <div className="text-[11px] font-semibold text-text-secondary">{title}</div>
+      <div className={`mt-1 text-sm font-bold ${toneClass.text}`}>{value}</div>
+      <div className="mt-1 text-xs leading-5 text-text-muted">{detail}</div>
+    </div>
+  );
+}
+
+function ThemePulseRow({ theme, rank }: { theme: ThemeCatalystScore; rank: number }) {
+  const score = Math.max(0, Math.min(theme.catalyst_score, 100));
+  const tone = score >= 70 ? "hot" : score >= 45 ? "warm" : "quiet";
+  const toneClass = getToneClass(tone);
+
+  return (
+    <div className="rounded-xl border border-border-subtle bg-surface-1/70 px-3 py-3">
+      <div className="flex items-start justify-between gap-3">
+        <div className="min-w-0 flex gap-3">
+          <div className={`flex h-7 w-7 shrink-0 items-center justify-center rounded-lg border font-mono text-xs ${toneClass.box} ${toneClass.text}`}>
+            {rank}
+          </div>
+          <div className="min-w-0">
+            <div className="text-sm font-semibold text-text-primary">{theme.theme_name}</div>
+            <div className="mt-1 text-[11px] text-text-muted line-clamp-1">
+              {theme.top_reasons?.[0] ?? "等待更多舆情证据"}
+            </div>
+          </div>
+        </div>
+        <div className={`font-mono text-sm font-bold tabular-nums ${toneClass.text}`}>{score.toFixed(0)}</div>
+      </div>
+      <div className="mt-3 h-1.5 rounded-full bg-surface-2 overflow-hidden">
+        <div className={`h-full rounded-full ${toneClass.bar}`} style={{ width: `${score}%` }} />
+      </div>
+    </div>
+  );
+}
+
+function NewsRow({ item, event }: { item: CatalystNewsItem; event?: CatalystEvent }) {
+  const status = getStatusMeta(item.status);
+  const eventTone = event ? getEventTone(event) : "quiet";
+  const toneClass = getToneClass(eventTone);
+  const title = event?.title || item.title;
+  const href = item.url || event?.url || "";
+
+  return (
+    <article className="rounded-2xl border border-border-subtle bg-surface-1/70 p-4 transition-colors hover:border-amber-500/15">
+      <div className="flex flex-col gap-3 md:flex-row md:items-start md:justify-between">
+        <div className="min-w-0">
+          <div className="flex flex-wrap items-center gap-2">
+            <span className={`rounded-md border px-2 py-0.5 text-[10px] ${status.className}`}>{status.label}</span>
+            <span className="text-[10px] text-text-muted">{formatSource(item.source)}</span>
+            <span className="text-[10px] text-text-muted">{formatDateTime(item.published_at || item.created_at)}</span>
+          </div>
+          {href ? (
+            <a href={href} target="_blank" rel="noreferrer" className="mt-2 block text-sm font-semibold leading-6 text-text-primary hover:text-amber-300">
+              {title}
+            </a>
+          ) : (
+            <h3 className="mt-2 text-sm font-semibold leading-6 text-text-primary">{title}</h3>
+          )}
+          {event?.summary || event?.llm_reason ? (
+            <p className="mt-2 text-xs leading-6 text-text-secondary line-clamp-2">
+              {event.llm_reason || event.summary}
+            </p>
+          ) : item.error ? (
+            <p className="mt-2 text-xs leading-6 text-amber-300/80 line-clamp-2">{item.error}</p>
+          ) : null}
+        </div>
+
+        <div className="grid grid-cols-3 gap-2 md:w-[240px] shrink-0">
+          <MiniScore label="强度" value={event?.strength} tone={eventTone} />
+          <MiniScore label="新鲜" value={event?.freshness} tone={eventTone} />
+          <MiniScore label="可信" value={event?.confidence} tone={eventTone} />
+        </div>
+      </div>
+      {event?.themes ? (
+        <div className="mt-3 rounded-xl border border-border-subtle bg-surface-2/70 px-3 py-2 text-[11px] leading-5 text-text-secondary">
+          归因：{event.themes}
+        </div>
+      ) : null}
+      {event ? (
+        <div className="mt-3 flex flex-wrap gap-2">
+          <span className={`rounded-md border px-2 py-1 text-[10px] ${toneClass.box} ${toneClass.text}`}>{formatCatalystType(event.catalyst_type)}</span>
+          <span className="rounded-md border border-border-subtle bg-surface-2 px-2 py-1 text-[10px] text-text-muted">本地已归档</span>
+        </div>
+      ) : null}
+    </article>
+  );
+}
+
+function CatalystInsight({ event }: { event: CatalystEvent }) {
+  const tone = getEventTone(event);
+  const toneClass = getToneClass(tone);
+  return (
+    <div className="rounded-xl border border-border-subtle bg-surface-1/70 px-3 py-3">
+      <div className="flex items-start justify-between gap-3">
+        <div className="min-w-0">
+          <div className="flex flex-wrap items-center gap-2">
+            <span className={`rounded-md border px-2 py-0.5 text-[10px] ${toneClass.box} ${toneClass.text}`}>{formatCatalystType(event.catalyst_type)}</span>
+            <span className="text-[10px] text-text-muted">{formatDateTime(event.published_at || event.created_at)}</span>
+          </div>
+          <div className="mt-2 text-sm font-semibold leading-6 text-text-primary line-clamp-2">{event.title}</div>
+          <div className="mt-1 text-xs leading-5 text-text-secondary line-clamp-3">
+            {event.llm_reason || event.summary || "等待更多解释"}
+          </div>
+        </div>
+        <div className={`font-mono text-sm font-bold tabular-nums ${toneClass.text}`}>{Math.round(event.strength)}</div>
+      </div>
+    </div>
+  );
+}
+
+function MiniScore({ label, value, tone }: { label: string; value?: number; tone: CatalystTone }) {
+  const toneClass = getToneClass(tone);
+  return (
+    <div className="rounded-lg bg-surface-2 px-2 py-1.5">
+      <div className="text-[10px] text-text-muted">{label}</div>
+      <div className={`mt-0.5 font-mono text-xs font-semibold tabular-nums ${value == null ? "text-text-muted" : toneClass.text}`}>
+        {value == null ? "--" : Math.round(value)}
+      </div>
+    </div>
+  );
+}
+
+function BoundaryLine({ text }: { text: string }) {
+  return (
+    <div className="flex items-start gap-2">
+      <span className="mt-2 h-1.5 w-1.5 shrink-0 rounded-full bg-amber-400" />
+      <span>{text}</span>
+    </div>
+  );
+}
+
+function EmptyLine({ text }: { text: string }) {
+  return (
+    <div className="rounded-xl border border-border-subtle bg-surface-1/70 px-3 py-4 text-center text-xs text-text-muted">
+      {text}
+    </div>
+  );
+}
+
+function getToneClass(tone: CatalystTone) {
+  if (tone === "hot") {
+    return {
+      box: "border-red-500/15 bg-red-500/[0.08]",
+      text: "text-red-300",
+      bar: "bg-red-400",
+    };
+  }
+  if (tone === "warm") {
+    return {
+      box: "border-amber-500/15 bg-amber-500/[0.08]",
+      text: "text-amber-300",
+      bar: "bg-amber-400",
+    };
+  }
+  return {
+    box: "border-border-subtle bg-surface-2",
+    text: "text-text-secondary",
+    bar: "bg-text-muted",
+  };
+}
+
+function getEventTone(event: CatalystEvent): CatalystTone {
+  if (event.strength >= 70 || event.confidence >= 75) return "hot";
+  if (event.strength >= 45 || event.freshness >= 55) return "warm";
+  return "quiet";
+}
+
+function getStatusMeta(status: string) {
+  if (status === "analyzed") {
+    return { label: "已归因", className: "border-emerald-500/15 bg-emerald-500/10 text-emerald-300" };
+  }
+  if (status === "pending") {
+    return { label: "待处理", className: "border-amber-500/15 bg-amber-500/10 text-amber-300" };
+  }
+  if (status === "failed") {
+    return { label: "异常", className: "border-red-500/15 bg-red-500/10 text-red-300" };
+  }
+  if (status === "skipped") {
+    return { label: "已忽略", className: "border-border-subtle bg-surface-2 text-text-muted" };
+  }
+  return { label: status || "未知", className: "border-border-subtle bg-surface-2 text-text-muted" };
+}
+
+function formatCatalystType(type: string) {
+  const labels: Record<string, string> = {
+    policy: "政策",
+    industry: "产业",
+    event: "事件",
+    earnings: "业绩",
+    announcement: "公告",
+    news: "新闻",
+  };
+  return labels[type] ?? type;
+}
+
+function formatSource(source: string) {
+  if (!source) return "未知来源";
+  return source.replace("tushare:", "").replace("rss:", "");
+}
+
+function formatDateTime(value?: string | null) {
+  if (!value) return "暂无时间";
+  const date = new Date(value);
+  if (Number.isNaN(date.getTime())) return value;
+  return date.toLocaleString("zh-CN", {
+    month: "2-digit",
+    day: "2-digit",
+    hour: "2-digit",
+    minute: "2-digit",
+  });
+}
diff --git a/frontend/src/app/(auth)/stock/[code]/page.tsx b/frontend/src/app/(auth)/stock/[code]/page.tsx
index d462b95a..3dbe2993 100644
--- a/frontend/src/app/(auth)/stock/[code]/page.tsx
+++ b/frontend/src/app/(auth)/stock/[code]/page.tsx
@@ -107,7 +107,7 @@ export default function StockDetailPage() {
 
         const recFromHistory = history
           .flatMap((group) => group.recommendations)
-          .find((rec) => rec.ts_code === code);
+          .find((rec) => sameStockCode(rec.ts_code, code));
         const rec = thesisData?.recommendation ?? recFromHistory ?? null;
 
         if (rec) {
@@ -122,11 +122,6 @@ export default function StockDetailPage() {
           setRecScore(null);
         }
 
-        setQuote(null);
-        setSignals(null);
-        setKline([]);
-        setCapitalFlow([]);
-        setEvidenceLoaded(false);
       } finally {
         if (!cancelled) setLoading(false);
       }
@@ -138,35 +133,52 @@ export default function StockDetailPage() {
     };
   }, [code]);
 
-  const loadEvidence = async () => {
-    if (!code || evidenceLoading || evidenceLoaded) return;
-    setEvidenceLoading(true);
-    try {
-      const [quoteData, signalsData, klineData, flowData] = await Promise.all([
-        fetchAPI<QuoteData>(`/api/stocks/${code}/quote`).catch(() => null),
-        fetchAPI<StockSignals>(`/api/stocks/${code}/signals`).catch(() => null),
-        fetchAPI<unknown[]>(`/api/stocks/${code}/kline?days=120`).catch(() => []),
-        fetchAPI<FlowRecord[]>(`/api/stocks/${code}/capital_flow?days=10`).catch(() => []),
-      ]);
-      setQuote(isValidQuote(quoteData) ? quoteData : null);
-      setSignals(signalsData);
-      setKline(Array.isArray(klineData) ? klineData : []);
-      setCapitalFlow(Array.isArray(flowData) ? flowData : []);
-      setEvidenceLoaded(true);
-    } finally {
-      setEvidenceLoading(false);
+  useEffect(() => {
+    if (!code) return;
+
+    let cancelled = false;
+
+    async function loadEvidence() {
+      setEvidenceLoading(true);
+      setEvidenceLoaded(false);
+      setQuote(null);
+      setSignals(null);
+      setKline([]);
+      setCapitalFlow([]);
+
+      try {
+        const [quoteData, signalsData, klineData, flowData] = await Promise.all([
+          fetchAPI<QuoteData>(`/api/stocks/${code}/quote`).catch(() => null),
+          fetchAPI<StockSignals>(`/api/stocks/${code}/signals`).catch(() => null),
+          fetchAPI<unknown[]>(`/api/stocks/${code}/kline?days=120`).catch(() => []),
+          fetchAPI<FlowRecord[]>(`/api/stocks/${code}/capital_flow?days=10`).catch(() => []),
+        ]);
+
+        if (cancelled) return;
+        setQuote(isValidQuote(quoteData) ? quoteData : null);
+        setSignals(signalsData);
+        setKline(Array.isArray(klineData) ? klineData : []);
+        setCapitalFlow(Array.isArray(flowData) ? flowData : []);
+        setEvidenceLoaded(true);
+      } finally {
+        if (!cancelled) setEvidenceLoading(false);
+      }
     }
-  };
+
+    loadEvidence();
+    return () => {
+      cancelled = true;
+    };
+  }, [code]);
 
   const recommendation = thesis?.recommendation;
   const latestTracking = thesis?.latest_tracking;
   const latestFlow = capitalFlow.length > 0 ? capitalFlow[capitalFlow.length - 1] : null;
   const pageName = recommendation?.name || thesis?.name || quote?.name || code;
-  const conviction = recommendation?.llm_score != null ? Math.round(recommendation.llm_score) : null;
 
   return (
     <ErrorBoundary>
-      <div className="max-w-7xl mx-auto px-4 md:px-8 pt-6 pb-20 md:pb-10 space-y-5">
+      <div className="max-w-7xl mx-auto px-4 md:px-6 pt-4 pb-20 md:pb-8 space-y-4">
         <a
           href="/recommendations"
           className="inline-flex items-center gap-1.5 text-xs text-text-muted hover:text-text-primary transition-colors animate-fade-in-up"
@@ -177,12 +189,11 @@ export default function StockDetailPage() {
           返回推荐池
         </a>
 
-        <div className="glass-card-static p-5 animate-fade-in-up overflow-hidden relative">
-          <div className="absolute right-[-90px] top-[-120px] w-72 h-72 rounded-full bg-cyan-500/[0.04] blur-3xl pointer-events-none" />
-          <div className="relative grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_320px] gap-5">
+        <div className="glass-card-static p-4 md:p-5 animate-fade-in-up overflow-hidden relative">
+          <div className="relative grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_360px] gap-4">
             <div className="min-w-0">
               <div className="flex flex-wrap items-center gap-2 mb-2">
-                <span className="text-[10px] uppercase tracking-[0.22em] text-cyan-400 font-semibold">个股作战卡</span>
+                <span className="text-[10px] tracking-[0.18em] text-cyan-400 font-semibold">个股作战卡</span>
                 {recommendation?.action_plan ? (
                   <span className={`text-[10px] px-2 py-0.5 rounded-full border ${getActionPlanClass(recommendation.action_plan)}`}>
                     {recommendation.action_plan}
@@ -197,33 +208,34 @@ export default function StockDetailPage() {
 
               <div className="flex flex-wrap items-end gap-3">
                 <h1 className="text-2xl md:text-3xl font-bold tracking-tight">{pageName}</h1>
-                <span className="text-sm text-text-muted font-mono tabular-nums">{code}</span>
+                <span className="text-sm text-text-muted font-mono tabular-nums">{quote?.ts_code || thesis?.ts_code || code}</span>
               </div>
 
-              <p className="text-sm text-text-secondary leading-relaxed mt-3 max-w-4xl">
+              <p className="text-sm text-text-secondary leading-relaxed mt-2 max-w-4xl">
                 {buildHeroSummary(thesis, quote)}
               </p>
 
-              <div className="grid grid-cols-1 md:grid-cols-3 gap-2 mt-4">
-                {(thesis?.decision_points ?? []).slice(0, 3).map((point) => (
-                  <DecisionPoint key={point.label} label={point.label} value={point.value} />
-                ))}
+              <div className="grid grid-cols-1 md:grid-cols-3 gap-2 mt-3">
+                <DecisionPoint label="触发条件" value={recommendation?.trigger_condition || "等待触发条件归档"} />
+                <DecisionPoint label="失效条件" value={recommendation?.invalidation_condition || "等待失效条件归档"} />
+                <DecisionPoint label="仓位建议" value={recommendation?.suggested_position_pct != null ? `${recommendation.suggested_position_pct}%` : "未设置"} />
               </div>
             </div>
 
-            <div className="rounded-2xl bg-surface-1/80 border border-border-subtle p-4">
-              <SectionTitle title="今日处理" />
-              <div className="text-xs text-text-secondary leading-relaxed mt-2">
-                {thesis ? "已读取最近推荐、跟踪和诊断记录。" : "加载中"}
+            <div className="rounded-xl bg-surface-1/80 border border-border-subtle p-4">
+              <div className="flex items-center justify-between gap-3">
+                <SectionTitle title="实时盘面" />
+                <EvidenceStatus loading={evidenceLoading} loaded={evidenceLoaded} hasQuote={!!quote} />
               </div>
+              <QuotePriceLine quote={quote} loading={evidenceLoading && !quote} />
               <div className="grid grid-cols-2 gap-2 mt-3">
-                <MiniDataCell label="当前动作" value={recommendation?.action_plan || "观察"} />
-                <MiniDataCell label="把握度" value={conviction != null ? `${conviction}/10` : "暂无"} />
-                <MiniDataCell label="初步判断" value={formatPrefilterDecision(recommendation?.prefilter_decision)} />
-                <MiniDataCell label="建议仓位" value={recommendation?.suggested_position_pct != null ? `${recommendation.suggested_position_pct}%` : "未设置"} />
+                <MiniDataCell label="换手" value={quote?.turnover_rate != null ? `${quote.turnover_rate.toFixed(2)}%` : evidenceLoading ? "加载中" : "暂无"} />
+                <MiniDataCell label="量比" value={quote?.volume_ratio != null ? quote.volume_ratio.toFixed(2) : evidenceLoading ? "加载中" : "暂无"} />
+                <MiniDataCell label="主力净流" value={latestFlow ? formatFlowAmount(latestFlow.main_net_inflow) : evidenceLoading ? "加载中" : "暂无"} tone={latestFlow?.main_net_inflow} />
+                <MiniDataCell label="位置" value={signals ? positionComment(recScore?.position_score ?? signals.position_score) : evidenceLoading ? "加载中" : "暂无"} />
               </div>
               {(recommendation?.recall_tags?.length ?? 0) > 0 ? (
-                <div className="flex flex-wrap gap-2 mt-3">
+                <div className="flex flex-wrap gap-1.5 mt-3">
                   {(recommendation?.recall_tags ?? []).slice(0, 4).map((tag) => (
                     <span key={tag} className="text-[10px] px-2 py-1 rounded-md bg-surface-2 text-text-muted border border-border-subtle">
                       {formatRecallTag(tag)}
@@ -250,45 +262,24 @@ export default function StockDetailPage() {
           <div className="glass-card-static p-8 text-center text-sm text-text-muted">加载个股推演中...</div>
         ) : (
           <>
-            <div className="grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_320px] gap-4 animate-fade-in-up">
+            <div className="grid grid-cols-1 xl:grid-cols-[minmax(0,1fr)_360px] gap-4 animate-fade-in-up">
               <div className="space-y-4">
+                <EvidenceCard recommendation={recommendation} quote={quote} signals={signals} loading={evidenceLoading} />
+                <div className="grid grid-cols-1 lg:grid-cols-2 gap-4">
+                  {kline.length > 0 ? <KlineChart data={kline as never[]} /> : <ChartEmptyCard title="价格走势" description={evidenceLoading ? "行情证据加载中" : "暂无K线数据"} />}
+                  {capitalFlow.length > 0 ? <CapitalFlowChart data={capitalFlow} /> : <ChartEmptyCard title="资金趋势" description={evidenceLoading ? "资金数据加载中" : "暂无资金流数据"} />}
+                </div>
+                {latestFlow ? <CapitalFlowBreakdown flow={latestFlow} /> : null}
                 <PlanCard recommendation={recommendation} trackingNote={latestTracking?.review_note || ""} />
-                <EvidenceCard recommendation={recommendation} quote={quote} signals={signals} />
+              </div>
+
+              <div className="space-y-4">
+                <SignalSnapshot signals={signals} recScore={recScore} evidenceLoading={evidenceLoading} evidenceLoaded={evidenceLoaded} />
+                <QuoteSnapshot quote={quote} evidenceLoading={evidenceLoading} evidenceLoaded={evidenceLoaded} />
+                <TrackingCard tracking={latestTracking ?? null} />
                 <DiagnosisArchiveCard diagnoses={thesis?.diagnoses ?? []} />
               </div>
-
-              <div className="space-y-4">
-                <TrackingCard tracking={latestTracking ?? null} />
-                <QuoteSnapshot quote={quote} evidenceLoaded={evidenceLoaded} />
-                <SignalSnapshot signals={signals} recScore={recScore} evidenceLoaded={evidenceLoaded} />
-              </div>
             </div>
-
-            <div className="glass-card-static p-4 animate-fade-in-up">
-              <div className="flex flex-wrap items-center justify-between gap-3">
-                <SectionTitle title="行情证据" />
-                <button
-                  onClick={loadEvidence}
-                  disabled={evidenceLoading || evidenceLoaded}
-                  className="text-xs px-3 py-2 rounded-xl border border-border-subtle bg-surface-1 text-text-secondary hover:text-cyan-400 transition-colors disabled:opacity-40"
-                >
-                  {evidenceLoading ? "加载中..." : evidenceLoaded ? "已加载证据" : "加载行情证据"}
-                </button>
-              </div>
-            </div>
-
-            {evidenceLoaded ? (
-              <>
-                <div className="grid grid-cols-1 md:grid-cols-2 gap-4 animate-fade-in-up">
-                  {kline.length > 0 ? <KlineChart data={kline as never[]} /> : <ChartEmptyCard title="K线图" description="暂无K线数据" />}
-                  {capitalFlow.length > 0 ? <CapitalFlowChart data={capitalFlow} /> : <ChartEmptyCard title="资金流向趋势" description="暂无资金流数据" />}
-                </div>
-
-                {latestFlow ? (
-                  <CapitalFlowBreakdown flow={latestFlow} />
-                ) : null}
-              </>
-            ) : null}
           </>
         )}
       </div>
@@ -307,12 +298,11 @@ function PlanCard({
     <div className="glass-card-static p-5">
       <div className="flex items-center justify-between gap-3 mb-3">
         <SectionTitle title="执行计划" />
-        {recommendation?.llm_score != null ? (
-          <span className="text-xs font-mono tabular-nums text-cyan-400/80">把握度 {Math.round(recommendation.llm_score)}/10</span>
+        {recommendation?.score != null ? (
+          <span className="text-xs font-mono tabular-nums text-cyan-400/80">规则分 {Math.round(recommendation.score)}</span>
         ) : null}
       </div>
       <div className="space-y-3 text-sm">
-        {recommendation?.prefilter_reason ? <PlanRow label="初筛理由" value={recommendation.prefilter_reason} /> : null}
         {(recommendation?.focus_points?.length ?? 0) > 0 ? (
           <PlanRow label="关注点" value={(recommendation?.focus_points ?? []).slice(0, 3).join(" / ")} />
         ) : null}
@@ -328,6 +318,44 @@ function PlanCard({
   );
 }
 
+function EvidenceStatus({ loading, loaded, hasQuote }: { loading: boolean; loaded: boolean; hasQuote: boolean }) {
+  const text = loading ? "同步中" : loaded && hasQuote ? "已同步" : loaded ? "暂无行情" : "等待同步";
+  const color = loading
+    ? "border-amber-500/20 bg-amber-500/10 text-amber-400"
+    : loaded && hasQuote
+      ? "border-cyan-500/20 bg-cyan-500/10 text-cyan-400"
+      : "border-border-subtle bg-surface-2 text-text-muted";
+
+  return <span className={`text-[10px] px-2 py-0.5 rounded-full border ${color}`}>{text}</span>;
+}
+
+function QuotePriceLine({ quote, loading }: { quote: QuoteData | null; loading: boolean }) {
+  if (!quote) {
+    return (
+      <div className="mt-3 rounded-xl bg-surface-2 px-3 py-3 text-sm text-text-muted">
+        {loading ? "正在读取最新行情证据..." : "暂无实时行情"}
+      </div>
+    );
+  }
+
+  return (
+    <div className="flex items-end justify-between gap-3 mt-3">
+      <div>
+        <div className="text-[10px] text-text-muted">现价</div>
+        <div className={`text-4xl font-bold font-mono tabular-nums leading-none ${quote.pct_chg > 0 ? "text-red-400" : quote.pct_chg < 0 ? "text-emerald-400" : "text-text-primary"}`}>
+          {quote.price.toFixed(2)}
+        </div>
+      </div>
+      <div className="text-right">
+        <div className="text-[10px] text-text-muted">涨跌幅</div>
+        <div className={`text-lg font-bold font-mono tabular-nums ${quote.pct_chg > 0 ? "text-red-400" : quote.pct_chg < 0 ? "text-emerald-400" : "text-text-muted"}`}>
+          {quote.pct_chg > 0 ? "+" : ""}{quote.pct_chg.toFixed(2)}%
+        </div>
+      </div>
+    </div>
+  );
+}
+
 function formatRecallTag(tag: string): string {
   const labels: Record<string, string> = {
     sector_recall: "主线入选",
@@ -341,23 +369,16 @@ function formatRecallTag(tag: string): string {
   return labels[tag] ?? tag;
 }
 
-function formatPrefilterDecision(decision?: string | null): string {
-  const labels: Record<string, string> = {
-    priority: "优先研究",
-    watch: "保留观察",
-    ignore: "暂不处理",
-  };
-  return labels[decision ?? ""] ?? "暂无";
-}
-
 function EvidenceCard({
   recommendation,
   quote,
   signals,
+  loading,
 }: {
   recommendation: RecommendationData | null | undefined;
   quote: QuoteData | null;
   signals: StockSignals | null;
+  loading: boolean;
 }) {
   const reasons = recommendation?.reasons ?? [];
   const evidenceChips = [
@@ -369,12 +390,15 @@ function EvidenceCard({
     signals?.pullback_support ? "回踩支撑" : null,
   ].filter(Boolean) as string[];
   return (
-    <div className="glass-card-static p-5">
-      <SectionTitle title="作战依据" />
+    <div className="glass-card-static p-4 md:p-5">
+      <div className="flex items-center justify-between gap-3">
+        <SectionTitle title="核心依据" />
+        {loading ? <span className="text-[10px] text-amber-400">行情同步中</span> : null}
+      </div>
       {evidenceChips.length ? (
         <div className="flex flex-wrap gap-2 mt-3">
           {evidenceChips.slice(0, 6).map((item) => (
-            <span key={item} className="text-[11px] px-2.5 py-1 rounded-lg bg-surface-2 border border-border-subtle text-text-secondary">
+            <span key={item} className="text-[11px] px-2.5 py-1 rounded-md bg-surface-2 border border-border-subtle text-text-secondary">
               {item}
             </span>
           ))}
@@ -391,7 +415,7 @@ function EvidenceCard({
         )}
       </div>
       {recommendation?.risk_note ? (
-        <div className="mt-4 rounded-xl bg-amber-500/[0.04] border border-amber-500/10 px-3 py-2 text-xs text-amber-400/80 leading-relaxed">
+        <div className="mt-4 rounded-lg bg-amber-500/[0.04] border border-amber-500/10 px-3 py-2 text-xs text-amber-400/80 leading-relaxed">
           风险提示：{recommendation.risk_note}
         </div>
       ) : null}
@@ -447,10 +471,21 @@ function TrackingCard({ tracking }: { tracking: StockThesisResponse["latest_trac
   );
 }
 
-function QuoteSnapshot({ quote, evidenceLoaded }: { quote: QuoteData | null; evidenceLoaded: boolean }) {
+function QuoteSnapshot({
+  quote,
+  evidenceLoading,
+  evidenceLoaded,
+}: {
+  quote: QuoteData | null;
+  evidenceLoading: boolean;
+  evidenceLoaded: boolean;
+}) {
   return (
     <div className="glass-card-static p-5">
-      <SectionTitle title="盘面表现" />
+      <div className="flex items-center justify-between gap-3">
+        <SectionTitle title="盘口细节" />
+        <EvidenceStatus loading={evidenceLoading} loaded={evidenceLoaded} hasQuote={!!quote} />
+      </div>
       {quote ? (
         <>
           <div className="flex items-baseline gap-3 mt-3">
@@ -471,7 +506,7 @@ function QuoteSnapshot({ quote, evidenceLoaded }: { quote: QuoteData | null; evi
       ) : evidenceLoaded ? (
         <div className="text-sm text-text-muted mt-3">暂无行情证据</div>
       ) : (
-        <div className="text-sm text-text-muted mt-3">点击“加载行情证据”后查看</div>
+        <div className="text-sm text-text-muted mt-3">{evidenceLoading ? "正在读取行情证据..." : "等待行情同步"}</div>
       )}
     </div>
   );
@@ -480,10 +515,12 @@ function QuoteSnapshot({ quote, evidenceLoaded }: { quote: QuoteData | null; evi
 function SignalSnapshot({
   signals,
   recScore,
+  evidenceLoading,
   evidenceLoaded,
 }: {
   signals: StockSignals | null;
   recScore: RecScore | null;
+  evidenceLoading: boolean;
   evidenceLoaded: boolean;
 }) {
   return (
@@ -506,7 +543,7 @@ function SignalSnapshot({
       ) : evidenceLoaded ? (
         <div className="text-sm text-text-muted mt-3">暂无技术信号</div>
       ) : (
-        <div className="text-sm text-text-muted mt-3">点击“加载行情证据”后查看</div>
+        <div className="text-sm text-text-muted mt-3">{evidenceLoading ? "正在计算技术信号..." : "等待信号同步"}</div>
       )}
     </div>
   );
@@ -537,7 +574,7 @@ function CapitalFlowBreakdown({ flow }: { flow: FlowRecord }) {
 
 function ChartEmptyCard({ title, description }: { title: string; description: string }) {
   return (
-    <div className="bg-bg-card rounded-xl p-4">
+    <div className="bg-bg-card rounded-xl p-4 border border-border-subtle/70">
       <h2 className="text-sm font-medium text-text-secondary mb-2">{title}</h2>
       <div className="h-56 flex items-center justify-center text-sm text-text-muted">{description}</div>
     </div>
@@ -546,8 +583,8 @@ function ChartEmptyCard({ title, description }: { title: string; description: st
 
 function DecisionPoint({ label, value }: { label: string; value: string }) {
   return (
-    <div className="rounded-xl bg-surface-1/70 border border-border-subtle px-3 py-2.5">
-      <div className="text-[10px] text-text-muted uppercase tracking-wider font-semibold">{label}</div>
+    <div className="rounded-lg bg-surface-1/70 border border-border-subtle px-3 py-2.5">
+      <div className="text-[10px] text-text-muted tracking-wider font-semibold">{label}</div>
       <div className="text-sm text-text-primary leading-relaxed mt-1">{value}</div>
     </div>
   );
@@ -562,11 +599,18 @@ function PlanRow({ label, value }: { label: string; value: string }) {
   );
 }
 
-function MiniDataCell({ label, value }: { label: string; value: string | number }) {
+function MiniDataCell({ label, value, tone }: { label: string; value: string | number; tone?: number | null }) {
+  const toneClass = tone == null
+    ? "text-text-secondary"
+    : tone > 0
+      ? "text-red-400"
+      : tone < 0
+        ? "text-emerald-400"
+        : "text-text-secondary";
   return (
-    <div className="rounded-xl bg-surface-2 px-3 py-2">
+    <div className="rounded-lg bg-surface-2 px-3 py-2">
       <div className="text-[10px] text-text-muted/60">{label}</div>
-      <div className="text-sm font-mono tabular-nums text-text-secondary mt-0.5">{value}</div>
+      <div className={`text-sm font-mono tabular-nums mt-0.5 ${toneClass}`}>{value}</div>
     </div>
   );
 }
@@ -613,7 +657,7 @@ function FlowBar({ label, value, max }: { label: string; value: number; max: num
 }
 
 function SectionTitle({ title }: { title: string }) {
-  return <div className="text-[10px] text-text-muted uppercase tracking-wider font-semibold">{title}</div>;
+  return <div className="text-[10px] text-text-muted tracking-wider font-semibold">{title}</div>;
 }
 
 function getLifecycleLabel(status: string) {
@@ -626,7 +670,7 @@ function getLifecycleLabel(status: string) {
     expired: "到期复盘",
     invalidated: "已失效",
   };
-  return labels[status] ?? status;
+  return labels[status] ?? "观察";
 }
 
 function getActionPlanClass(actionPlan: string) {
@@ -667,6 +711,8 @@ function signalTypeLabel(signalType?: string) {
     launch: "启动",
     reversal: "反转",
     breakout_confirm: "突破确认",
+    flow_momentum: "资金推动",
+    none: "观察",
   };
   return map[signalType || ""] ?? "观察";
 }
@@ -684,3 +730,11 @@ function positionComment(positionScore: number) {
   if (positionScore >= 50) return "位置中性";
   return "位置偏高，防追高";
 }
+
+function sameStockCode(left: string, right: string) {
+  return stripMarket(left) === stripMarket(right);
+}
+
+function stripMarket(value: string) {
+  return value.replace(/\.(SH|SZ|BJ)$/i, "").replace(/^(SH|SZ|BJ)/i, "");
+}
diff --git a/frontend/src/components/nav.tsx b/frontend/src/components/nav.tsx
index bcee9f0c..56dc4784 100644
--- a/frontend/src/components/nav.tsx
+++ b/frontend/src/components/nav.tsx
@@ -34,6 +34,17 @@ function FireIcon() {
   );
 }
 
+function RadarIcon() {
+  return (
+    <svg width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="1.8" strokeLinecap="round" strokeLinejoin="round">
+      <path d="M12 20a8 8 0 1 0-8-8" />
+      <path d="M12 16a4 4 0 1 0-4-4" />
+      <path d="M12 12l7-7" />
+      <path d="M4 20h16" />
+    </svg>
+  );
+}
+
 function StrategyIcon() {
   return (
     <svg width="18" height="18" viewBox="0 0 24 24" fill="none" stroke="currentColor" strokeWidth="1.8" strokeLinecap="round" strokeLinejoin="round">
@@ -117,6 +128,7 @@ export function SidebarNav() {
       <SideNavItem href="/dashboard" icon={<DashboardIcon />} label="今日作战" />
       <SideNavItem href="/recommendations" icon={<TargetIcon />} label="推荐池" />
       <SideNavItem href="/sectors" icon={<FireIcon />} label="板块主线" />
+      <SideNavItem href="/sentiment" icon={<RadarIcon />} label="舆情雷达" />
       <SideNavItem href="/watchlists" icon={<WatchlistIcon />} label="自选股" />
       <SideNavItem href="/chat" icon={<ChatIcon />} label="研究助手" />
       <SideNavItem href="/diagnose" icon={<DiagnoseIcon />} label="个股诊断" />
@@ -162,6 +174,9 @@ export function MobileBottomNav() {
         <MobileNavItem href="/chat" label="助手">
           <ChatIcon />
         </MobileNavItem>
+        <MobileNavItem href="/sentiment" label="舆情">
+          <RadarIcon />
+        </MobileNavItem>
         <MobileNavItem href="/watchlists" label="自选">
           <WatchlistIcon />
         </MobileNavItem>
diff --git a/frontend/src/components/stock-card.tsx b/frontend/src/components/stock-card.tsx
index 1c43e86f..7c6dc311 100644
--- a/frontend/src/components/stock-card.tsx
+++ b/frontend/src/components/stock-card.tsx
@@ -4,11 +4,10 @@ import type { RecommendationData } from "@/lib/api";
 
 export default function StockCard({ rec, compact = false }: { rec: RecommendationData; compact?: boolean }) {
   const action = getActionMeta(rec.action_plan);
-  const trigger = rec.trigger_condition ?? rec.entry_timing ?? rec.decision_trace?.headline ?? rec.prefilter_reason ?? rec.reasons?.[0] ?? "等待触发条件确认";
+  const trigger = rec.trigger_condition ?? rec.entry_timing ?? rec.decision_trace?.headline ?? rec.reasons?.[0] ?? "等待触发条件确认";
   const risk = rec.invalidation_condition ?? rec.risk_note ?? "暂无明确失效条件";
   const thesis = rec.decision_trace?.headline ?? rec.reasons?.[0] ?? rec.focus_points?.[0] ?? "等待更多盘面证据";
-  const conviction = rec.llm_score != null ? Math.round(rec.llm_score) : null;
-  const chips = buildChips(rec, conviction).slice(0, compact ? 3 : 5);
+  const chips = buildChips(rec).slice(0, compact ? 3 : 5);
 
   return (
     <a
@@ -85,7 +84,7 @@ function getActionMeta(actionPlan?: string | null) {
   return { label: "观察", className: "border-border-subtle bg-surface-2 text-text-muted" };
 }
 
-function buildChips(rec: RecommendationData, conviction: number | null) {
+function buildChips(rec: RecommendationData) {
   const recallLabels: Record<string, string> = {
     sector_recall: "主线入选",
     trend_scan: "趋势入选",
@@ -100,7 +99,7 @@ function buildChips(rec: RecommendationData, conviction: number | null) {
   return [
     rec.entry_signal_type ? signalTypeLabel(rec.entry_signal_type) : null,
     rec.review_after_days ? `${rec.review_after_days}日复盘` : null,
-    conviction != null ? `把握 ${conviction}/10` : null,
+    rec.score != null ? `规则分 ${Math.round(rec.score)}` : null,
     ...(rec.recall_tags ?? []).map((tag) => recallLabels[tag] ?? tag),
   ].filter(Boolean) as string[];
 }
diff --git a/frontend/src/lib/api.ts b/frontend/src/lib/api.ts
index d1d69984..9c36d3f1 100644
--- a/frontend/src/lib/api.ts
+++ b/frontend/src/lib/api.ts
@@ -242,6 +242,45 @@ export interface SectorData {
   catalyst_reasons?: string[];
 }
 
+export interface CatalystNewsItem {
+  id: number;
+  title: string;
+  source: string;
+  url: string;
+  published_at: string | null;
+  status: "pending" | "analyzed" | "skipped" | "failed" | string;
+  catalyst_id: number | null;
+  error: string;
+  created_at: string;
+  analyzed_at: string | null;
+}
+
+export interface CatalystEvent {
+  id: number;
+  title: string;
+  summary: string;
+  source: string;
+  url: string;
+  published_at: string | null;
+  catalyst_type: "policy" | "industry" | "event" | "earnings" | "announcement" | "news" | string;
+  strength: number;
+  freshness: number;
+  confidence: number;
+  raw_text: string;
+  llm_reason: string;
+  created_at: string;
+  themes?: string;
+}
+
+export interface ThemeCatalystScore {
+  theme_id: string;
+  theme_name: string;
+  catalyst_score: number;
+  catalyst_count: number;
+  top_reasons: string[];
+  generated_by: string;
+}
+
 export interface LatestResult {
   market_temperature: MarketTemperatureData | null;
   recommendations: RecommendationData[];