aria-code 0.167.0__py3-none-win_amd64.whl

aria_code/apps/cli/commands/market_cmds.py

@@ -0,0 +1,1419 @@
+"""
+MarketCommandsMixin — Market commands: quote, realty, football, screen, news, screen_cn, limitup, north.
+
+Extracted from aria_cli.py. Methods' __globals__ are rebound to aria_cli's namespace
+by _rebind_mixin_globals() called at module load time.
+"""
+
+from __future__ import annotations
+
+from aria_code.apps.cli.i18n import ui_text
+
+from ._ui import print_error
+from typing import Optional, Tuple
+
+
+_FOOTBALL_CONNECTORS = (
+    "对阵", "对战", "对决", " vs ", " VS ", "vs", "VS", " v.s. ",
+    " versus ", "跟", "和", "与", "对", "pk", "PK",
+)
+
+_FOOTBALL_STRONG_INTENT_TERMS = (
+    "比分", "比赛预测", "比赛", "对阵", "交手", "胜负", "几比几",
+    "进球", "足球", "国家队", "世界杯", "欧洲杯", "欧冠", "英超",
+    "西甲", "德甲", "意甲", "法甲", "中超", "美职联",
+    "打败", "战胜", "击败", "打平", "晋级", "出线", "夺冠", "踢",
+    "score", "match", "football", "soccer", "beat",
+)
+
+_FOOTBALL_AMBIGUOUS_INTENT_TERMS = (
+    "预测", "谁赢", "谁能赢", "谁会赢", "结果预测",
+    "predict", "prediction", "win",
+)
+
+_MARKET_CONTEXT_TERMS = (
+    "股票", "股价", "成交量", "市值", "行情", "k线", "K线", "图表",
+    "技术指标", "均线", "支撑", "阻力", "涨跌", "涨幅", "跌幅",
+    "财报", "财务", "估值", "营收", "利润", "持仓", "风险", "基金",
+    "ETF", "etf", "期权", "债券", "外汇", "期货", "RSI", "MACD",
+    "买入", "卖出", "做多", "做空", "quote", "stock", "share",
+    "equity", "volume", "market cap", "earnings", "revenue", "price",
+)
+
+
+def _rss_items_from_xml(xml_text: str, limit: int = 5) -> list[dict]:
+    """Parse simple RSS item fields without external dependencies."""
+    import html as _html
+    import xml.etree.ElementTree as _ET
+
+    try:
+        root = _ET.fromstring(xml_text)
+    except Exception:
+        return []
+    items: list[dict] = []
+    for item in root.findall(".//item"):
+        title = item.findtext("title") or ""
+        link = item.findtext("link") or ""
+        pub_date = item.findtext("pubDate") or item.findtext("published") or ""
+        source = item.findtext("source") or ""
+        if not title:
+            continue
+        items.append({
+            "title": _html.unescape(title.strip()),
+            "url": link.strip(),
+            "published_at": pub_date.strip(),
+            "source": source.strip() or "RSS",
+        })
+        if len(items) >= limit:
+            break
+    return items
+
+
+def _fetch_public_news_fallback(topic: str, limit: int = 5) -> list[dict]:
+    """Fetch public RSS news without API keys.
+
+    Yahoo Finance works well for tickers; Google News RSS covers private
+    companies such as SpaceX. This is a best-effort fallback, not a guaranteed
+    research source.
+    """
+    import re as _re
+    import urllib.parse as _parse
+    import urllib.request as _request
+
+    topic = (topic or "market").strip()
+    urls: list[str] = []
+    if _re.match(r"^[A-Z]{1,6}(?:[.-][A-Z]{1,3})?$", topic):
+        urls.append(
+            "https://feeds.finance.yahoo.com/rss/2.0/headline?"
+            f"s={_parse.quote(topic)}&region=US&lang=en-US"
+        )
+    query = f"{topic} latest news when:14d"
+    urls.append(
+        "https://news.google.com/rss/search?"
+        f"q={_parse.quote(query)}&hl=en-US&gl=US&ceid=US:en"
+    )
+    headers = {"User-Agent": "Mozilla/5.0 AriaCode/4.1"}
+    for url in urls:
+        try:
+            req = _request.Request(url, headers=headers)
+            with _request.urlopen(req, timeout=8) as resp:
+                text = resp.read().decode("utf-8", errors="replace")
+            items = _rss_items_from_xml(text, limit=limit)
+            if items:
+                return items
+        except Exception:
+            continue
+    return []
+
+
+def _is_known_football_name(name: str) -> bool:
+    """Return True only when a fragment resolves to a known football team/country."""
+    n = (name or "").strip()
+    if not n:
+        return False
+    try:
+        from football_data_client import _CN_TEAM_MAP, _FIFA_RATINGS
+    except Exception:
+        return False
+    if n in _CN_TEAM_MAP:
+        return True
+    nl = n.lower()
+    for cn, en in _CN_TEAM_MAP.items():
+        if n == cn or nl == str(en).lower():
+            return True
+    for en_key, data in _FIFA_RATINGS.items():
+        if nl == str(en_key).lower() or n == str(data.get("name", "")):
+            return True
+    return False
+
+
+def _is_probable_football_query(text: str, pair: Optional[Tuple[str, str]] = None) -> bool:
+    """Guard the NL football route so finance queries do not enter Poisson mode."""
+    raw = text or ""
+    if not raw.strip() or raw.strip().startswith("/"):
+        return False
+    if any(term in raw for term in _MARKET_CONTEXT_TERMS):
+        return False
+    pair = pair or _parse_nl_team_pair(raw)
+    if not pair:
+        return False
+    known_pair = _is_known_football_name(pair[0]) and _is_known_football_name(pair[1])
+    if any(term in raw for term in _FOOTBALL_STRONG_INTENT_TERMS):
+        return True
+    if any(term in raw for term in _FOOTBALL_AMBIGUOUS_INTENT_TERMS):
+        return known_pair
+    if not any(conn.lower() in raw.lower() for conn in _FOOTBALL_CONNECTORS):
+        return False
+    return known_pair
+
+
+def _parse_nl_team_pair(text: str) -> Optional[Tuple[str, str]]:
+    """
+    Extract (home_cn, away_cn) from a natural-language football query.
+
+    Handles patterns like:
+      "葡萄牙和刚果比赛比分预测"
+      "巴西跟阿根廷谁赢"
+      "英格兰对阵法国"
+      "Germany vs France prediction"    ← English also supported
+    Returns None if two teams cannot be confidently identified.
+    """
+    try:
+        from football_data_client import _CN_TEAM_MAP, _FIFA_RATINGS
+    except Exception:
+        return None
+
+    # Build reverse map: english_lower → cn_name (from _CN_TEAM_MAP values)
+    _EN_TO_CN: dict = {}
+    for cn, en in _CN_TEAM_MAP.items():
+        _EN_TO_CN.setdefault(en.lower(), cn)
+    # Also add direct FIFA rating keys → cn name
+    for en_key, data in _FIFA_RATINGS.items():
+        cn_name = data.get("name", "")
+        if cn_name and en_key.lower() not in _EN_TO_CN:
+            _EN_TO_CN[en_key.lower()] = cn_name
+
+    # Ordered connectors — longer ones first to avoid partial matches
+    _CONNECTORS = _FOOTBALL_CONNECTORS
+    # Words to strip from team-name fragments
+    _STRIP_WORDS = (
+        "预测", "分析", "比赛", "比分", "胜率", "结果", "谁赢", "谁会赢",
+        "今天", "今日", "明天", "的", "了", "吗", "呢",
+        "prediction", "match", "game", "preview", "who wins", "predict",
+        "football", "soccer", "score",
+    )
+
+    def _clean(s: str) -> str:
+        s = s.strip("？！，。、《》（）[]【】:：'\"-— \t")
+        for w in sorted(_STRIP_WORDS, key=len, reverse=True):
+            s = s.replace(w, "").strip()
+        return s.strip()
+
+    def _resolve(name: str) -> Optional[str]:
+        """Resolve a name (CN or EN) to its canonical Chinese name."""
+        name = name.strip()
+        if not name:
+            return None
+        # Direct CN lookup
+        if name in _CN_TEAM_MAP:
+            return name
+        # English → CN
+        nl = name.lower()
+        if nl in _EN_TO_CN:
+            return _EN_TO_CN[nl]
+        # Partial English match
+        for en_key, cn_n in _EN_TO_CN.items():
+            if nl in en_key or en_key in nl:
+                return cn_n
+        # Partial CN match
+        for cn in _CN_TEAM_MAP:
+            if name in cn or cn in name:
+                return cn
+        # Return as-is if it looks like a real name (≥2 chars)
+        return name if len(name) >= 2 else None
+
+    # ── Approach 1: split on connector ───────────────────────────────────────
+    for conn in _CONNECTORS:
+        if conn.lower() in text.lower():
+            idx = text.lower().index(conn.lower())
+            left  = _resolve(_clean(text[:idx]))
+            right = _resolve(_clean(text[idx + len(conn):]))
+            if left and right and left != right:
+                return left, right
+
+    # ── Approach 2: scan for all known Chinese team names in text order ──────
+    found: list = []
+    for cn in _CN_TEAM_MAP:
+        if cn in text:
+            found.append((text.index(cn), cn))
+    # Also scan English names (word-boundary, case-insensitive)
+    import re as _re
+    tl = text.lower()
+    for en_key, cn_n in _EN_TO_CN.items():
+        if len(en_key) < 3:
+            continue
+        m = _re.search(r'\b' + _re.escape(en_key) + r'\b', tl)
+        if m:
+            found.append((m.start(), cn_n))
+    found.sort()
+    # Remove duplicates keeping earlier occurrence
+    seen_en: set = set()
+    unique: list = []
+    for pos, cn in found:
+        en = _CN_TEAM_MAP.get(cn, cn)
+        if en not in seen_en:
+            seen_en.add(en)
+            unique.append((pos, cn))
+    if len(unique) >= 2:
+        return unique[0][1], unique[1][1]
+
+    return None
+
+
+import logging
+import json
+import asyncio
+import datetime
+import time
+import shlex
+from typing import Dict, Any, Optional
+
+def print_quote_result(*args, **kwargs):
+    from aria_cli import print_quote_result as fn
+    return fn(*args, **kwargs)
+def _football_fixtures(*args, **kwargs):
+    from aria_cli import _football_fixtures as fn
+    return fn(*args, **kwargs)
+# execute_aria_tool lives in apps/cli/tool_executor.py and needs nothing from
+# aria_cli — the cache it reads is owned there now, so the hop was pointless.
+from aria_code.apps.cli.tool_executor import execute_aria_tool
+def _is_ashare_symbol(*args, **kwargs):
+    from aria_code.apps.cli.helpers import _is_ashare_symbol as fn
+    return fn(*args, **kwargs)
+def _prompt_str(*args, **kwargs):
+    from aria_cli import _prompt_str as fn
+    return fn(*args, **kwargs)
+def _render_reits_list(*args, **kwargs):
+    from aria_cli import _render_reits_list as fn
+    return fn(*args, **kwargs)
+def _render_multi_city(*args, **kwargs):
+    from aria_cli import _render_multi_city as fn
+    return fn(*args, **kwargs)
+def _football_team(*args, **kwargs):
+    from aria_cli import _football_team as fn
+    return fn(*args, **kwargs)
+def _get_LOCAL_TOOLS():
+    # Owned by apps/cli/tool_registry.py; aria_cli fills it in place.
+    from ..tool_registry import LOCAL_TOOLS as val
+    return val
+def _render_property_val(*args, **kwargs):
+    from aria_cli import _render_property_val as fn
+    return fn(*args, **kwargs)
+def _get_mdc(*args, **kwargs):
+    from .._optional import get_mdc as fn
+    return fn(*args, **kwargs)
+def _load_data_keys(*args, **kwargs):
+    from aria_code.apps.cli.provider_keys import load_data_keys as fn
+    return fn(*args, **kwargs)
+def _render_asset_score(*args, **kwargs):
+    from aria_cli import _render_asset_score as fn
+    return fn(*args, **kwargs)
+def _render_rental_yield(*args, **kwargs):
+    from aria_cli import _render_rental_yield as fn
+    return fn(*args, **kwargs)
+def _football_h2h(*args, **kwargs):
+    from aria_cli import _football_h2h as fn
+    return fn(*args, **kwargs)
+def _football_standings(*args, **kwargs):
+    from aria_cli import _football_standings as fn
+    return fn(*args, **kwargs)
+def _render_house_price(*args, **kwargs):
+    from aria_cli import _render_house_price as fn
+    return fn(*args, **kwargs)
+def format_quote_output(*args, **kwargs):
+    from aria_cli import format_quote_output as fn
+    return fn(*args, **kwargs)
+def _prompt_float(*args, **kwargs):
+    from aria_cli import _prompt_float as fn
+    return fn(*args, **kwargs)
+def _ashare_code_to_name(*args, **kwargs):
+    from aria_cli import _ashare_code_to_name as fn
+    return fn(*args, **kwargs)
+def _get__HAS_MDC():
+    from .._optional import HAS_MDC as val
+    return val
+def parse_symbols(*args, **kwargs):
+    from aria_cli import parse_symbols as fn
+    return fn(*args, **kwargs)
+
+import json
+import asyncio
+import datetime
+import time
+import shlex
+import sys
+import os
+from typing import Dict, Any, Optional
+
+
+import json
+import asyncio
+import datetime
+import time
+import shlex
+import sys
+import os
+from typing import Dict, Any, Optional
+
+# A real logger. This was a `def logger(*args, **kwargs)` forwarding to
+# aria_cli, so every `logger.debug(...)` raised AttributeError — and all but
+# one of them sit in an except block, where the failure replaced whatever
+# error was being reported.
+logger = logging.getLogger(__name__)
+
+
+def _news_date(raw: str) -> str:
+    """A feed's publish time as 2026-10-05.
+
+    The first ten characters were kept, which suits ISO dates and cut an RSS
+    date mid-word: "Mon, 05 Oct 2026 12:00:00 GMT" showed as "Mon, 05 Oc".
+    """
+    from datetime import datetime
+    from email.utils import parsedate_to_datetime
+
+    text = str(raw or "").strip()
+    for parse in (lambda t: datetime.fromisoformat(t.replace("Z", "+00:00")), parsedate_to_datetime):
+        try:
+            return parse(text).strftime("%Y-%m-%d")
+        except (TypeError, ValueError, IndexError):
+            continue
+    return text[:10]
+
+
+class MarketCommandsMixin:
+    """Mixin: Market commands: quote, realty, football, screen, news, screen_cn, limitup, north."""
+
+    async def cmd_realty(self, args: str):
+        """
+        /realty market [city1] [city2]  — 城市房价指数
+        /realty reit [code]             — REIT 列表或单只分析
+        /realty valuation               — 物业估值计算器（交互式）
+        /realty rent                    — 租金收益率计算（交互式）
+        /realty compare [cities...]     — 多城市对比
+        /realty score                   — 资产区位评分（交互式）
+        /realty us                      — 美国住房数据
+        """
+        import asyncio as _asyncio
+        loop = _asyncio.get_event_loop()
+        parts = args.strip().split() if args.strip() else []
+        sub = parts[0].lower() if parts else "market"
+
+        try:
+            from realty_data_tools import (
+                get_house_price_index, get_re_investment,
+                get_reits_list, get_reit_analysis, get_multi_city_comparison,
+                calc_rental_yield, property_valuation, asset_location_score,
+                get_us_housing_data,
+            )
+        except ImportError as e:
+            if self.context.has_rich:
+                self.context.console.print(f"[red]realty_data_tools 未加载: {e}[/red]")
+            return
+
+        if sub == "market":
+            city1 = parts[1] if len(parts) > 1 else "北京"
+            city2 = parts[2] if len(parts) > 2 else ("上海" if city1 != "上海" else "北京")
+            import functools as _functools
+            if self.context.has_rich:
+                with self.context.console.status(f"[dim]{ui_text(self, f'获取 {city1}/{city2} 房价指数', f'Fetching {city1}/{city2} house price indices')}...[/dim]", spinner="dots"):
+                    r = await loop.run_in_executor(
+                        None, _functools.partial(get_house_price_index, city1, city2)
+                    )
+            else:
+                r = get_house_price_index(city1, city2)
+            _render_house_price(r)
+            # Also show investment data
+            if self.context.has_rich:
+                with self.context.console.status(f"[dim]{ui_text(self, '获取房地产投资数据', 'Fetching real-estate investment data')}...[/dim]", spinner="dots"):
+                    ri = await loop.run_in_executor(None, get_re_investment)
+            else:
+                ri = get_re_investment()
+            if ri.get("success") and ri.get("latest"):
+                lt = ri["latest"]
+                if self.context.has_rich:
+                    self.context.console.print(f"\n  [dim]房地产开发投资[/dim]  {lt.get('日期','')}  "
+                                  f"最新值 [bold]{lt.get('最新值','')}[/bold]  "
+                                  f"涨跌 {lt.get('涨跌幅','')}  "
+                                  f"近1年 {lt.get('近1年涨跌幅','')}")
+
+        elif sub == "reit":
+            code = parts[1] if len(parts) > 1 else None
+            if code:
+                if self.context.has_rich:
+                    with self.context.console.status(f"[dim]{ui_text(self, f'分析 {code} REIT', f'Analyzing REIT {code}')}...[/dim]", spinner="dots"):
+                        r = await loop.run_in_executor(None, get_reit_analysis, code)
+                else:
+                    r = get_reit_analysis(code)
+                if r.get("success"):
+                    if self.context.has_rich:
+                        self.context.console.print(f"\n  [bold cyan]{r.get('code','')}[/bold cyan] "
+                                      f"[dim]{r.get('name','')}[/dim]")
+                        self.context.console.print(f"  现价 [bold]{r.get('price','')}[/bold]  "
+                                      f"涨跌 {r.get('chg_pct','')}%")
+                        if r.get("return_1y") is not None:
+                            rc = "green" if r["return_1y"] > 0 else "red"
+                            self.context.console.print(f"  近1年收益: [{rc}]{r['return_1y']:+.2f}%[/{rc}]")
+                        if r.get("volatility_annual"):
+                            self.context.console.print(f"  年化波动率: {r['volatility_annual']:.2f}%")
+                else:
+                    self.context.console.print(f"[red]{r.get('error','分析失败')}[/red]") if self.context.has_rich else None
+            else:
+                if self.context.has_rich:
+                    with self.context.console.status(f"[dim]{ui_text(self, '获取 REIT 列表', 'Fetching the REIT list')}...[/dim]", spinner="dots"):
+                        r = await loop.run_in_executor(None, get_reits_list)
+                else:
+                    r = get_reits_list()
+                _render_reits_list(r)
+
+        elif sub == "compare":
+            cities = parts[1:] if len(parts) > 1 else None
+            if self.context.has_rich:
+                with self.context.console.status(f"[dim]{ui_text(self, '对比多城市房价', 'Comparing city house prices')}...[/dim]", spinner="dots"):
+                    r = await loop.run_in_executor(None, get_multi_city_comparison, cities)
+            else:
+                r = get_multi_city_comparison(cities)
+            _render_multi_city(r)
+
+        elif sub in ("rent", "rental"):
+            if self.context.has_rich:
+                self.context.console.print("[bold]💰 租金收益率计算器[/bold]  [dim](输入 0 跳过可选项)[/dim]")
+            price_wan  = _prompt_float("购入价格(万元): ", 200.0)
+            monthly_rent = _prompt_float("月租金(元): ", 5000.0)
+            annual_costs = _prompt_float("年维护成本(元)[可选]: ", 0.0)
+            loan_ratio = _prompt_float("贷款成数 0-1 (如0.7=七成)[可选]: ", 0.0)
+            p = {"purchase_price": price_wan, "monthly_rent": monthly_rent,
+                 "annual_costs": annual_costs, "loan_ratio": loan_ratio}
+            r = calc_rental_yield(p)
+            _render_rental_yield(r)
+
+        elif sub in ("valuation", "val"):
+            if self.context.has_rich:
+                self.context.console.print("[bold]🏢 物业估值计算器[/bold]")
+            area = _prompt_float("建筑面积(㎡): ", 100.0)
+            monthly_rent = _prompt_float("月租金(元): ", 5000.0)
+            tier = _prompt_str("区位层级 (tier1/tier2/tier3): ", "tier2")
+            p = {"area_sqm": area, "monthly_rent": monthly_rent, "location_tier": tier}
+            r = property_valuation(p)
+            _render_property_val(r)
+
+        elif sub == "score":
+            if self.context.has_rich:
+                self.context.console.print("[bold]📍 资产区位评分[/bold]")
+            city  = _prompt_str("城市: ", "上海")
+            area  = _prompt_float("建筑面积(㎡): ", 100.0)
+            floor_n = int(_prompt_float("楼层: ", 1.0))
+            traffic = _prompt_str("客流量 (high/medium/low): ", "medium")
+            fire_ok = _prompt_str("允许明火? (y/n): ", "n").lower() in ("y","yes","是")
+            reno_ok = _prompt_str("允许改造? (y/n): ", "y").lower() in ("y","yes","是")
+            p = {"city": city, "area_sqm": area, "floor": floor_n,
+                 "foot_traffic": traffic, "open_fire_allowed": fire_ok,
+                 "renovation_allowed": reno_ok}
+            r = asset_location_score(p)
+            _render_asset_score(r)
+
+        elif sub == "us":
+            if self.context.has_rich:
+                with self.context.console.status(f"[dim]{ui_text(self, '获取美国住房数据', 'Fetching US housing data')}...[/dim]", spinner="dots"):
+                    r = await loop.run_in_executor(None, get_us_housing_data)
+            else:
+                r = get_us_housing_data()
+            if not r.get("success"):
+                if self.context.has_rich: self.context.console.print(f"[red]{r.get('error')}[/red]")
+                return
+            if self.context.has_rich:
+                from rich.table import Table as _T
+                from rich import box as _box
+                tb = _T(title="[bold]🏠 美国住房市场数据[/bold]", box=_box.ROUNDED)
+                tb.add_column("指标", style="dim"); tb.add_column("最新值"); tb.add_column("日期", style="dim")
+                for key, val in r.get("data", {}).items():
+                    lt = val.get("latest", {})
+                    v = lt.get("value")
+                    tb.add_row(val.get("label", key), str(v) if v else "—", str(lt.get("date",""))[:7])
+                self.context.console.print(tb)
+                for line in r.get("assessment", []):
+                    self.context.console.print(f"  [dim]▸ {line}[/dim]")
+
+        else:
+            if self.context.has_rich:
+                self.context.console.print("[dim]用法: /realty [market|reit|compare|rent|valuation|score|us][/dim]")
+                self.context.console.print("[dim]示例: /realty market 北京 上海[/dim]")
+                self.context.console.print("[dim]      /realty reit 508603[/dim]")
+                self.context.console.print("[dim]      /realty rent  (交互式租金计算)[/dim]")
+                self.context.console.print("[dim]      /realty compare 北京 上海 成都 杭州[/dim]")
+
+    async def cmd_football(self, args: str):
+        """
+        足球赛事分析和预测
+
+        子命令:
+          /football standings <联赛>              联赛积分榜
+          /football fixtures  <联赛> [days]       近期赛程（默认7天）
+          /football predict   <主队> vs <客队> [联赛]  比赛预测
+          /football team      <球队名> [联赛]      球队近期状态
+          /football h2h       <队1> vs <队2> [联赛]   历史交锋
+
+        联赛代码: pl/epl/英超  bl/德甲  ll/西甲  sa/意甲  fl1/法甲  cl/欧冠
+        示例:
+          /football standings pl
+          /football predict Arsenal vs Chelsea pl
+          /football team Manchester City pl
+          /football fixtures cl 14
+        """
+        from rich.table import Table
+        from rich import box as rich_box
+        from rich.panel import Panel
+
+        parts = args.strip().split()
+        if not parts:
+            self.context.console.print(Panel(
+                "[bold]足球分析命令[/bold]\n\n"
+                "  [cyan]/football standings pl[/cyan]              英超积分榜\n"
+                "  [cyan]/football fixtures cl 14[/cyan]            欧冠未来14天赛程\n"
+                "  [cyan]/football predict Arsenal vs Chelsea[/cyan] 预测比赛结果\n"
+                "  [cyan]/football team Bayern Munich bl[/cyan]     球队近期状态\n"
+                "  [cyan]/football h2h Barcelona vs Real Madrid[/cyan] 历史交锋\n\n"
+                "[dim]联赛: pl/英超  bl/德甲  ll/西甲  sa/意甲  fl1/法甲  cl/欧冠[/dim]\n"
+                "[dim]需要设置 FOOTBALL_DATA_API_KEY（football-data.org 免费注册）[/dim]",
+                title="[bold]⚽ Football Analyst[/bold]",
+                border_style="green",
+            ))
+            return
+
+        sub = parts[0].lower()
+
+        # ── standings ──────────────────────────────────────────────────────────
+        if sub == "standings":
+            league = parts[1] if len(parts) > 1 else "pl"
+            await self._run_in_executor(_football_standings, league)
+
+        # ── fixtures ──────────────────────────────────────────────────────────
+        elif sub in ("fixtures", "schedule", "赛程"):
+            league  = parts[1] if len(parts) > 1 else "pl"
+            days    = int(parts[2]) if len(parts) > 2 and parts[2].isdigit() else 7
+            await self._run_in_executor(_football_fixtures, league, days)
+
+        # ── predict ───────────────────────────────────────────────────────────
+        elif sub in ("predict", "预测", "prediction"):
+            raw = " ".join(parts[1:])
+            if " vs " in raw.lower():
+                idx     = raw.lower().index(" vs ")
+                home    = raw[:idx].strip()
+                rest    = raw[idx + 4:].strip()
+                away_parts = rest.split()
+                # last token might be league code (including wc/世界杯)
+                from football_data_client import LEAGUE_IDS, TOURNAMENT_CODES
+                _all_codes = {**LEAGUE_IDS, **{k: v for k, v in TOURNAMENT_CODES.items()}}
+                if away_parts and away_parts[-1].lower().replace(" ", "") in _all_codes:
+                    league = away_parts[-1]
+                    away   = " ".join(away_parts[:-1])
+                elif away_parts and away_parts[-1].lower() in ("wc", "worldcup", "世界杯", "ca", "ec"):
+                    league = away_parts[-1]
+                    away   = " ".join(away_parts[:-1])
+                else:
+                    league = "pl"
+                    away   = rest
+                await self._football_predict(home, away, league)
+            else:
+                self.context.console.print("[red]用法: /football predict <主队> vs <客队> [联赛/wc][/red]")
+
+        # ── team ──────────────────────────────────────────────────────────────
+        elif sub in ("team", "球队"):
+            rest  = " ".join(parts[1:])
+            from football_data_client import LEAGUE_IDS
+            tokens = rest.split()
+            if tokens and tokens[-1].lower() in LEAGUE_IDS:
+                league = tokens[-1]
+                team   = " ".join(tokens[:-1])
+            else:
+                league = "pl"
+                team   = rest
+            await self._run_in_executor(_football_team, team, league)
+
+        # ── h2h ───────────────────────────────────────────────────────────────
+        elif sub in ("h2h", "历史", "对决"):
+            raw = " ".join(parts[1:])
+            if " vs " in raw.lower():
+                idx  = raw.lower().index(" vs ")
+                t1   = raw[:idx].strip()
+                rest = raw[idx + 4:].strip()
+                from football_data_client import LEAGUE_IDS
+                tokens = rest.split()
+                if tokens and tokens[-1].lower() in LEAGUE_IDS:
+                    league = tokens[-1]
+                    t2     = " ".join(tokens[:-1])
+                else:
+                    league = "pl"
+                    t2     = rest
+                await self._run_in_executor(_football_h2h, t1, t2, league)
+            else:
+                self.context.console.print("[red]用法: /football h2h <队1> vs <队2> [联赛][/red]")
+
+        else:
+            # NL intent: /football 预测加拿大跟波黑... or /football 分析...
+            full_args = args.strip()
+            _has_cn = any('一' <= c <= '鿿' for c in full_args)
+            _has_kw = any(k in full_args.lower() for k in (
+                "predict", "preview", "analyze", "analysis", "who wins",
+                "预测", "分析", "谁赢", "比分", "胜率", "谁先", "开球",
+            ))
+            if _has_cn or _has_kw:
+                # ── Step 1: Parse two team names from NL text ─────────────────
+                _nl_pair = _parse_nl_team_pair(full_args)
+                if _nl_pair:
+                    _h_cn, _a_cn = _nl_pair
+                    # Determine league: national teams → wc, club → pl default
+                    try:
+                        from football_data_client import _CN_TEAM_MAP, _find_fifa_rating
+                        _h_en = _CN_TEAM_MAP.get(_h_cn, _h_cn)
+                        _a_en = _CN_TEAM_MAP.get(_a_cn, _a_cn)
+                        _is_nat = bool(_find_fifa_rating(_h_en) or _find_fifa_rating(_a_en))
+                    except Exception:
+                        _is_nat = True
+                    _nl_league = "wc" if _is_nat else "pl"
+                    await self._football_predict(_h_cn, _a_cn, _nl_league)
+                    return
+
+                # ── Step 2: Fall back to get_sports_context_for_query ─────────
+                try:
+                    from football_data_client import get_sports_context_for_query
+                    _sports_ctx = get_sports_context_for_query(full_args)
+                except Exception:
+                    _sports_ctx = ""
+                if _sports_ctx:
+                    _has_quant = "量化预测" in _sports_ctx
+                    _title = "⚽ 赛事预测" if _has_quant else "⚽ 赛事数据"
+                    self.context.console.print(Panel(
+                        _sports_ctx,
+                        title=f"[bold]{_title}[/bold]",
+                        border_style="cyan" if _has_quant else "blue",
+                    ))
+                else:
+                    self.context.console.print(
+                        "[yellow]⚽ 未能解析队名。[/yellow]\n"
+                        "支持格式：\n"
+                        "  [cyan]/football predict 葡萄牙 vs 刚果 wc[/cyan]\n"
+                        "  [cyan]/football 葡萄牙和刚果比赛[/cyan]  （自动识别）"
+                    )
+            else:
+                self.context.console.print(f"[red]未知子命令: {sub}[/red]  使用 /football 查看帮助")
+
+    async def _football_predict(self, home: str, away: str, league: str):
+        """Run football match prediction with LLM analysis."""
+        from rich.panel import Panel
+        from rich.table import Table
+        from rich import box as rich_box
+        import types
+
+        self.context.console.print(f"[#57606a]⚽ 分析 {home} vs {away} ({league.upper()})…[/#57606a]")
+
+        # WC / national team prediction path
+        _wc_leagues = {"wc", "worldcup", "世界杯", "world_cup", "ca", "ec", "afc"}
+        _is_wc = league.lower().replace(" ", "") in _wc_leagues
+
+        if _is_wc:
+            try:
+                from football_data_client import (
+                    predict_wc_match,
+                    _find_fifa_rating,
+                    team_display_name,
+                    football_prediction_quality,
+                    football_quality_missing_labels,
+                )
+                raw = predict_wc_match(home, away, neutral_venue=True)
+                _h_cn = team_display_name(raw.get("home_name_cn", home), "zh")
+                _a_cn = team_display_name(raw.get("away_name_cn", away), "zh")
+                # Build strength facts for display
+                _h_rank = raw.get("home_ranking", "?")
+                _a_rank = raw.get("away_ranking", "?")
+                _h_elo  = raw.get("home_elo")
+                _a_elo  = raw.get("away_elo")
+                _h_atk  = raw.get("home_attack")
+                _a_atk  = raw.get("away_attack")
+                _h_def  = raw.get("home_defense")
+                _a_def  = raw.get("away_defense")
+                _h_form = raw.get("home_form", "")
+                _a_form = raw.get("away_form", "")
+                _cal    = raw.get("calibrated_matches", 0)
+
+                def _rank_label(value):
+                    return f"#{value}" if value not in (None, "", "?") else "排名缺失"
+
+                _strength_facts = [
+                    f"FIFA排名: {_h_cn} {_rank_label(_h_rank)} · {_a_cn} {_rank_label(_a_rank)}",
+                ]
+                if _h_elo and _a_elo:
+                    _strength_facts.append(f"Elo评分: {_h_cn} {_h_elo:.0f} · {_a_cn} {_a_elo:.0f}")
+                if _h_atk is not None:
+                    _atk_h = f"{_h_atk:.2f}" if isinstance(_h_atk, float) else str(_h_atk)
+                    _atk_a = f"{_a_atk:.2f}" if isinstance(_a_atk, float) else str(_a_atk)
+                    _def_h = f"{_h_def:.2f}" if isinstance(_h_def, float) else str(_h_def)
+                    _def_a = f"{_a_def:.2f}" if isinstance(_a_def, float) else str(_a_def)
+                    _strength_facts.append(f"进攻强度: {_h_cn} {_atk_h} · {_a_cn} {_atk_a}")
+                    _strength_facts.append(f"防守强度: {_h_cn} {_def_h} · {_a_cn} {_def_a}")
+                _strength_facts.append(
+                    f"数据基础: {_cal} 场已完赛 WC 数据校准" if _cal > 0
+                    else "数据基础: FIFA排名 + Poisson引擎估算"
+                )
+                _quality = raw.get("data_quality") or football_prediction_quality(raw)
+                if _quality.get("missing"):
+                    _missing_labels = football_quality_missing_labels(_quality["missing"], "zh")
+                    _strength_facts.append(
+                        f"数据质量: {_quality.get('status', 'estimated')} · 缺失/估算 {', '.join(_missing_labels)}"
+                    )
+
+                pred = types.SimpleNamespace(
+                    home_win          = raw["home_win"],
+                    draw              = raw["draw"],
+                    away_win          = raw["away_win"],
+                    btts              = raw["btts"],
+                    lambda_home       = raw["lambda_home"],
+                    lambda_away       = raw["lambda_away"],
+                    most_likely       = raw["top_scorelines"][0]["score"] if raw["top_scorelines"] else "1-0",
+                    top_scores        = [{"score": s["score"], "prob": s["prob"]} for s in raw["top_scorelines"]],
+                    implied_odds      = raw["implied_odds"],
+                    key_factors       = _strength_facts,
+                    home_form         = _h_form,
+                    away_form         = _a_form,
+                    home_elo          = _h_elo,
+                    away_elo          = _a_elo,
+                    data_quality       = _quality,
+                    analysis          = "",
+                    verdict           = (
+                        f"[green]预测: {_h_cn} 获胜 ({raw['home_win']:.0%})[/green]" if raw["home_win"] > raw["away_win"] + 0.05
+                        else f"[green]预测: {_a_cn} 获胜 ({raw['away_win']:.0%})[/green]" if raw["away_win"] > raw["home_win"] + 0.05
+                        else f"[yellow]预测: 双方势均力敌，平局概率 {raw['draw']:.0%}[/yellow]"
+                    ),
+                    ht_lambda_home    = raw.get("ht_lambda_home", 0),
+                    ht_lambda_away    = raw.get("ht_lambda_away", 0),
+                    st_lambda_home    = raw.get("st_lambda_home", 0),
+                    st_lambda_away    = raw.get("st_lambda_away", 0),
+                    ht_home_win       = raw.get("ht_home_win", 0),
+                    ht_draw           = raw.get("ht_draw", 0),
+                    ht_away_win       = raw.get("ht_away_win", 0),
+                    ht_top_scorelines = raw.get("ht_top_scorelines", []),
+                    home_name_cn      = _h_cn,
+                    away_name_cn      = _a_cn,
+                )
+            except Exception as exc:
+                self.context.console.print(f"[red]WC 预测失败: {exc}[/red]")
+                return
+        else:
+            try:
+                from agents.sports.football_agent import FootballAgent
+
+                agent = FootballAgent(llm_call=None)
+
+                import asyncio
+                pred = await agent.predict(home, away, league, with_llm=False)
+
+                # Try LLM enhancement
+                if hasattr(self, 'terminal') and self.terminal:
+                    try:
+                        _score_candidates = "、".join(
+                            f"{s['score']}({s['prob']}%)"
+                            for s in getattr(pred, "top_scores", [])[:5]
+                        )
+                        llm_prompt = (
+                            f"你是专业足球分析师。简洁分析这场比赛（中文，不超过150字）:\n"
+                            f"{home} vs {away}\n"
+                            f"主队胜: {pred.home_win:.0%}  平: {pred.draw:.0%}  客队胜: {pred.away_win:.0%}\n"
+                            f"预期进球: {pred.lambda_home:.1f} - {pred.lambda_away:.1f}\n"
+                            f"最可能比分: {pred.most_likely}\n"
+                            f"候选比分（按概率降序）: {_score_candidates}\n"
+                            f"关键因素: {'; '.join(pred.key_factors)}\n"
+                            "规则: 只引用上方概率和候选比分，不要编造射正率、历史交锋、近5场客场等具体数据。"
+                        )
+                        analysis_text = await asyncio.wait_for(
+                            self.terminal._query_llm_async(llm_prompt),
+                            timeout=30
+                        )
+                        if analysis_text:
+                            pred.analysis = analysis_text
+                    except Exception:
+                        pass
+
+            except Exception as exc:
+                self.context.console.print(f"[red]预测失败: {exc}[/red]")
+                return
+
+        # ── 确定性分析文字：基于 Poisson 数字生成，不调用 LLM ────────────────
+        # 避免 gpt-oss:120b-cloud 忽略 enable_tools=False 并乱用工具
+        if not getattr(pred, "analysis", ""):
+            _h_n = getattr(pred, "home_name_cn", home)
+            _a_n = getattr(pred, "away_name_cn", away)
+            _hw  = pred.home_win
+            _dw  = pred.draw
+            _aw  = pred.away_win
+            _lh  = pred.lambda_home
+            _la  = pred.lambda_away
+            _ml  = pred.most_likely
+            # Determine favorite
+            if _hw > _aw + 0.08:
+                _tend = f"{_h_n} 胜算更大（{_hw:.0%}），为本场热门"
+            elif _aw > _hw + 0.08:
+                _tend = f"{_a_n} 胜算更大（{_aw:.0%}），为本场热门"
+            else:
+                _tend = f"双方势均力敌，{_h_n} 胜/平/负概率分别为 {_hw:.0%}/{_dw:.0%}/{_aw:.0%}"
+            # Expected goals narrative
+            _total = _lh + _la
+            if _total < 2.0:
+                _goal_desc = "预计进球偏少，防守型比赛"
+            elif _total < 3.0:
+                _goal_desc = "进球适中，攻防均衡"
+            else:
+                _goal_desc = "进球较多，进攻型对决"
+            pred.analysis = (
+                f"{_tend}。"
+                f"Poisson 模型预期进球 {_lh:.1f}–{_la:.1f}（共 {_total:.1f}），"
+                f"{_goal_desc}，最可能比分为 {_ml}。"
+            )
+
+        # ── display ──────────────────────────────────────────────────────────
+        from rich.columns import Columns
+        from rich.text import Text
+        try:
+            from football_data_client import team_display_name as _football_display_name
+        except Exception:
+            def _football_display_name(value, locale="zh"):
+                return str(value or "-")
+
+        # Probability bars
+        _home_display = _football_display_name(getattr(pred, "home_name_cn", home), "zh")
+        _away_display = _football_display_name(getattr(pred, "away_name_cn", away), "zh")
+
+        def pct_bar(val: float, width: int = 12) -> str:
+            filled = int(val * width)
+            return "█" * filled + "░" * (width - filled)
+
+        hw_color = "green" if pred.home_win > pred.away_win else "dim"
+        aw_color = "green" if pred.away_win > pred.home_win else "dim"
+        dw_color = "yellow" if pred.draw > 0.28 else "dim"
+
+        prob_table = Table(box=rich_box.SIMPLE, show_header=False, padding=(0, 1))
+        prob_table.add_column("", style="bold", width=16)
+        prob_table.add_column("", width=14)
+        prob_table.add_column("", width=6)
+        prob_table.add_column("", width=8)
+
+        prob_table.add_row(
+            f"[{hw_color}]{_home_display}[/{hw_color}]",
+            f"[{hw_color}]{pct_bar(pred.home_win)}[/{hw_color}]",
+            f"[{hw_color}]{pred.home_win:.0%}[/{hw_color}]",
+            f"[dim]赔率 {pred.implied_odds['home']}[/dim]",
+        )
+        prob_table.add_row(
+            f"[{dw_color}]平局[/{dw_color}]",
+            f"[{dw_color}]{pct_bar(pred.draw)}[/{dw_color}]",
+            f"[{dw_color}]{pred.draw:.0%}[/{dw_color}]",
+            f"[dim]赔率 {pred.implied_odds['draw']}[/dim]",
+        )
+        prob_table.add_row(
+            f"[{aw_color}]{_away_display}[/{aw_color}]",
+            f"[{aw_color}]{pct_bar(pred.away_win)}[/{aw_color}]",
+            f"[{aw_color}]{pred.away_win:.0%}[/{aw_color}]",
+            f"[dim]赔率 {pred.implied_odds['away']}[/dim]",
+        )
+
+        title = f"⚽ {_home_display} vs {_away_display}  [{league.upper()}]"
+        self.context.console.print(Panel(prob_table, title=f"[bold #3fb950]{title}[/bold #3fb950]", border_style="#3fb950"))
+
+        self.context.console.print(f"  [#57606a]预期进球: {_home_display} {pred.lambda_home:.2f} / {_away_display} {pred.lambda_away:.2f}"
+                      f"  │  双方均进球: {pred.btts:.0%}[/#57606a]")
+
+        # Top scorelines — show up to 8, colour-coded by outcome
+        _h_name = _home_display
+        _a_name = _away_display
+        if getattr(pred, "top_scores", None):
+            score_table = Table(box=rich_box.SIMPLE, show_header=True, padding=(0, 2))
+            score_table.add_column("可能比分", style="bold", width=12, no_wrap=True)
+            score_table.add_column("概率", justify="right", width=8, no_wrap=True)
+            score_table.add_column("结果", width=16, no_wrap=True)
+            for s in pred.top_scores[:8]:
+                _sc = s["score"]
+                _pr = s["prob"]
+                try:
+                    _hg, _ag = (_sc.split("-") + ["0"])[:2]
+                    _hg, _ag = int(_hg.strip()), int(_ag.strip())
+                except Exception:
+                    _hg, _ag = 0, 0
+                if _hg > _ag:
+                    _label = f"[#3fb950]{_h_name}胜[/#3fb950]"
+                    _sc_fmt = f"[green]{_sc}[/green]"
+                elif _ag > _hg:
+                    _label = f"[#f85149]{_a_name}胜[/#f85149]"
+                    _sc_fmt = f"[red]{_sc}[/red]"
+                else:
+                    _label = "[yellow]平局[/yellow]"
+                    _sc_fmt = f"[yellow]{_sc}[/yellow]"
+                score_table.add_row(_sc_fmt, f"{_pr}%", _label)
+            self.context.console.print(score_table)
+
+        # Half-time / second-half breakdown
+        if getattr(pred, "ht_lambda_home", 0) > 0:
+            _h_lbl = getattr(pred, "home_name_cn", home)
+            _h_lbl = _football_display_name(_h_lbl, "zh")
+            _a_lbl = _football_display_name(getattr(pred, "away_name_cn", away), "zh")
+            _ht_best = pred.ht_top_scorelines[0]["score"] if pred.ht_top_scorelines else "0-0"
+            _ht_best_p = pred.ht_top_scorelines[0]["prob"] if pred.ht_top_scorelines else 0
+            _st_best_lh = getattr(pred, "st_lambda_home", 0)
+            _st_best_la = getattr(pred, "st_lambda_away", 0)
+            self.context.console.print()
+            self.context.console.print(
+                f"  [bold]上半场[/bold]  预期进球 {_h_lbl} [cyan]{pred.ht_lambda_home:.2f}[/cyan] / "
+                f"{_a_lbl} [cyan]{pred.ht_lambda_away:.2f}[/cyan]"
+                f"  │  最可能: [bold]{_ht_best}[/bold] ({_ht_best_p}%)"
+            )
+            self.context.console.print(
+                f"  [#57606a]上半场胜/平/负: {pred.ht_home_win:.0%} / {pred.ht_draw:.0%} / {pred.ht_away_win:.0%}[/#57606a]"
+            )
+            _ht_scores_str = "  ".join(
+                f"[cyan]{s['score']}[/cyan] {s['prob']}%" for s in pred.ht_top_scorelines[:4]
+            )
+            if _ht_scores_str:
+                self.context.console.print(f"  [#57606a]比分分布: {_ht_scores_str}[/#57606a]")
+            self.context.console.print(
+                f"  [bold]下半场[/bold]  预期进球 {_h_lbl} [green]{_st_best_lh:.2f}[/green] / "
+                f"{_a_lbl} [green]{_st_best_la:.2f}[/green]"
+                f"  [#57606a](全场 − 上半场)[/#57606a]"
+            )
+            self.context.console.print()
+
+        # ── 实力对比 & 近期表现 ───────────────────────────────────────────────
+        _h_name_d = _home_display
+        _a_name_d = _away_display
+        _hform = getattr(pred, "home_form", "")
+        _aform = getattr(pred, "away_form", "")
+
+        if pred.key_factors:
+            self.context.console.print("\n  [bold]实力对比[/bold]")
+            for f_ in pred.key_factors:
+                self.context.console.print(f"  [#57606a]  • {f_}[/#57606a]")
+
+        # Form strings (W/D/L) from live API — only show when available
+        if _hform and _hform not in ("?????", ""):
+            def _form_colored(s: str) -> str:
+                out = []
+                for c in s.upper():
+                    if c == "W": out.append("[green]W[/green]")
+                    elif c == "D": out.append("[yellow]D[/yellow]")
+                    elif c == "L": out.append("[red]L[/red]")
+                    else: out.append(c)
+                return "".join(out)
+            self.context.console.print("\n  [bold]近期表现[/bold]")
+            self.context.console.print(f"  {_h_name_d}  {_form_colored(_hform)}")
+            if _aform and _aform not in ("?????", ""):
+                self.context.console.print(f"  {_a_name_d}  {_form_colored(_aform)}")
+
+        if pred.analysis:
+            self.context.console.print(Panel(
+                pred.analysis,
+                title="[bold]量化分析[/bold]",
+                border_style="#8c959f",
+                padding=(0, 2),
+            ))
+
+        self.context.console.print(f"\n  [bold green]{pred.verdict}[/bold green]")
+        self.context.console.print("  [#6e7781]提示：准确比分概率通常较分散，请按候选区间参考，不构成投注建议。[/#6e7781]\n")
+
+    async def cmd_screen(self, args: str):
+        """股票筛选: CN → screen_ashare; US → yfinance 大盘成分筛选."""
+        criteria = args.strip() or ""
+        low = criteria.lower()
+
+        # CN market detection
+        _cn_kw = ("a股", "沪深", "创业板", "科创板", "港股", "cn", "ashare", "沪市", "深市")
+        _is_cn = any(k in low for k in _cn_kw) or any(c.isdigit() for c in criteria[:6])
+
+        if _is_cn:
+            params: Dict[str, Any] = {}
+            for tok in args.split():
+                if "=" in tok:
+                    k, v = tok.split("=", 1)
+                    params[k.strip()] = v.strip()
+            if "screen_ashare" in _get_LOCAL_TOOLS():
+                await self._run_local_tool("screen_ashare", params, "A股选股筛选")
+            else:
+                await self.terminal.send_message(f"帮我筛选A股股票，条件：{criteria or '市值>50亿，非ST，流动性好'}")
+            return
+
+        # US / global: yfinance-based screening on a reference pool
+        import asyncio as _asyncio
+        _loop = _asyncio.get_event_loop()
+
+        _US_POOL = [
+            "AAPL","MSFT","NVDA","GOOGL","AMZN","META","TSLA","BRK-B","JPM","V",
+            "UNH","XOM","JNJ","WMT","MA","PG","LLY","HD","CVX","MRK",
+            "ABBV","PEP","KO","AVGO","COST","BAC","TMO","MCD","ACN","ADBE",
+            "CRM","NFLX","AMD","TXN","QCOM","INTC","CSCO","WFC","PM","VZ",
+            "RTX","HON","AMGN","LIN","DHR","UNP","CAT","SBUX","GS","BA",
+        ]
+
+        # Map common text criteria to filter presets
+        _growth_kw  = ("growth", "成长", "高增速", "tech", "科技", "ai", "人工智能")
+        _value_kw   = ("value", "价值", "低估", "dividend", "分红")
+        _momentum_kw= ("momentum", "动量", "趋势", "breakout", "突破")
+        _is_growth    = any(k in low for k in _growth_kw)
+        _is_value     = any(k in low for k in _value_kw)
+        _is_momentum  = any(k in low for k in _momentum_kw)
+
+        def _fetch_pool():
+            try:
+                import yfinance as _yf
+                tickers = _yf.Tickers(" ".join(_US_POOL))
+                rows = []
+                for sym in _US_POOL:
+                    try:
+                        info = tickers.tickers[sym].fast_info
+                        price     = getattr(info, "last_price", None) or 0
+                        mktcap    = getattr(info, "market_cap", None) or 0
+                        pe        = None   # filled below: fast_info has no P/E
+                        yr_return = getattr(info, "year_change", None)
+                        rows.append({
+                            "symbol": sym, "price": price,
+                            "mktcap": mktcap, "pe": pe,
+                            "yr_return": yr_return,
+                        })
+                    except Exception:
+                        pass
+                # fast_info has no pe_ratio, so P/E was blank for every row and
+                # the "value" screen (P/E 5-20) never matched anything. info has
+                # trailingPE; fetched in parallel, one request per symbol.
+                from concurrent.futures import ThreadPoolExecutor
+
+                def _pe(row):
+                    try:
+                        value = tickers.tickers[row["symbol"]].info.get("trailingPE")
+                        row["pe"] = float(value) if value else None
+                    except Exception:
+                        row["pe"] = None
+                with ThreadPoolExecutor(max_workers=10) as pool:
+                    list(pool.map(_pe, rows))
+                return rows
+            except Exception as _e:
+                logger.debug("screen US fetch error: %s", _e)
+                return []
+
+        if self.context.has_rich:
+            _status_msg = "[dim]" + ui_text(self, f"筛选 {len(_US_POOL)} 只美股 ({criteria or '按市值'})…",
+                                            f"Screening {len(_US_POOL)} US stocks ({criteria or 'by market cap'})…") + "[/dim]"
+            with self.context.console.status(_status_msg, spinner="dots"):
+                rows = await _loop.run_in_executor(None, _fetch_pool)
+        else:
+            print("  " + ui_text(self, "筛选美股中…", "Screening US stocks…"))
+            rows = await _loop.run_in_executor(None, _fetch_pool)
+
+        if not rows:
+            await self.terminal.send_message(
+                f"Screen US stocks matching: {criteria or 'large-cap'}. "
+                "Show top 10 with price, P/E, market cap, 1-year return."
+            )
+            return
+
+        # Apply simple filters
+        if _is_growth:
+            rows = [r for r in rows if (r.get("yr_return") or 0) > 0.15]
+        elif _is_value:
+            rows = [r for r in rows if r.get("pe") and 5 < r["pe"] < 20]
+        elif _is_momentum:
+            rows = sorted(rows, key=lambda r: r.get("yr_return") or 0, reverse=True)
+        else:
+            rows = sorted(rows, key=lambda r: r.get("mktcap") or 0, reverse=True)
+
+        rows = rows[:15]
+
+        if not rows:
+            msg = "[yellow]" + ui_text(self, f"当前条件 '{criteria}' 无匹配标的（池: {len(_US_POOL)} 只）",
+                                       f"Nothing matches '{criteria}' (pool: {len(_US_POOL)} stocks)") + "[/yellow]"
+            self.context.console.print(msg) if self.context.has_rich else print(msg.replace("[yellow]","").replace("[/yellow]",""))
+            return
+
+        if self.context.has_rich:
+            from rich.table import Table as _Tbl
+            t = _Tbl(title=ui_text(self, f"美股筛选  {criteria or 'large-cap'}  共 {len(rows)} 只",
+                                   f"US screen  {criteria or 'large-cap'}  {len(rows)} stocks"),
+                     show_header=True, box=None, padding=(0, 1))
+            t.add_column(ui_text(self, "代码", "Symbol"), style="bold", width=8)
+            t.add_column(ui_text(self, "价格", "Price"), justify="right")
+            t.add_column(ui_text(self, "市值(B$)", "Mkt cap ($B)"), justify="right", style="dim")
+            t.add_column(ui_text(self, "市盈率", "P/E (ttm)"), justify="right", style="dim")
+            t.add_column(ui_text(self, "年涨跌%", "1y %"), justify="right")
+            for r in rows:
+                yr  = r.get("yr_return")
+                yr_s = f"{yr*100:+.1f}%" if yr is not None else "—"
+                yr_color = "green" if (yr or 0) >= 0 else "red"
+                pe_s = f"{r['pe']:.1f}" if r.get("pe") and r["pe"] == r["pe"] else "—"
+                mc_s = f"{r['mktcap']/1e9:.0f}" if (r.get("mktcap") or 0) > 0 else "—"
+                t.add_row(
+                    r["symbol"],
+                    f"{r['price']:.2f}" if r.get("price") else "—",
+                    mc_s, pe_s,
+                    f"[{yr_color}]{yr_s}[/{yr_color}]",
+                )
+            self.context.console.print(t)
+            self.context.console.print("  [dim]" + ui_text(self, f"来源: yfinance · 池: {len(_US_POOL)} 只大市值美股",
+                                                          f"Source: yfinance · pool: {len(_US_POOL)} large-cap US stocks")
+                                       + "[/dim]")
+        else:
+            print(f"  美股筛选  {criteria}")
+            for r in rows:
+                yr = r.get("yr_return")
+                yr_s = f"{yr*100:+.1f}%" if yr is not None else "—"
+                print(f"  {r['symbol']:<8} ${r.get('price',0):.2f}  {yr_s}")
+
+    async def cmd_news(self, args: str):
+        """Fetch latest financial news for a topic or symbol.
+
+        Usage: /news [topic|symbol] [--limit N]
+        Examples:
+          /news AAPL
+          /news earnings --limit 10
+          /news crypto --limit 3
+        """
+        parts = args.split()
+        limit = 5
+        topic_parts = []
+        i = 0
+        while i < len(parts):
+            if parts[i] == "--limit" and i + 1 < len(parts):
+                try:
+                    limit = max(1, min(20, int(parts[i + 1])))
+                    i += 2
+                    continue
+                except ValueError:
+                    pass
+            topic_parts.append(parts[i])
+            i += 1
+        topic = " ".join(topic_parts) or "market"
+
+        self.context.console.print(f"[dim]Fetching {limit} news items for '{topic}'...[/dim]" if self.context.has_rich
+                      else f"Fetching news for {topic}...")
+
+        # Try backend first, then local tools (Finnhub / NewsAPI / AKShare fallback chain)
+        result = await execute_aria_tool(self.terminal.api_url, "analyze_news", {
+            "query": topic, "limit": limit,
+        })
+        if not result.get("success") and "analyze_news" in _get_LOCAL_TOOLS():
+            # Local fallback: uses Finnhub → NewsAPI → AKShare depending on configured keys
+            local_fn = _get_LOCAL_TOOLS()["analyze_news"][0]
+            result = await asyncio.get_event_loop().run_in_executor(
+                None, local_fn, {"query": topic, "symbol": topic, "limit": limit}
+            )
+        if result.get("success"):
+            data = result.get("data", {})
+            if isinstance(data, dict):
+                articles = data.get("articles", data.get("news", []))
+                sentiment = data.get("sentiment", data.get("overall_sentiment", ""))
+            elif isinstance(data, list):
+                articles = data
+                sentiment = ""
+            else:
+                articles = []
+                sentiment = ""
+            if not (isinstance(articles, list) and articles):
+                articles = await asyncio.get_event_loop().run_in_executor(
+                    None, _fetch_public_news_fallback, topic, limit
+                )
+                if articles:
+                    sentiment = "public RSS fallback"
+            if isinstance(articles, list) and articles:
+                if self.context.has_rich:
+                    self.context.console.print()
+                    if sentiment:
+                        sent_color = "green" if "positive" in sentiment.lower() or "bullish" in sentiment.lower() else (
+                            "red" if "negative" in sentiment.lower() or "bearish" in sentiment.lower() else "yellow"
+                        )
+                        self.context.console.print(f"  Sentiment: [{sent_color}]{sentiment}[/{sent_color}]")
+                        self.context.console.print()
+                for idx, a in enumerate(articles[:limit], 1):
+                    if isinstance(a, dict):
+                        title = a.get("title", "Untitled")
+                        source = a.get("source", a.get("publisher", ""))
+                        url_item = a.get("url", a.get("link", ""))
+                        pub_date = a.get("published_at", a.get("date", a.get("publishedAt", "")))
+                        if pub_date:
+                            pub_date = _news_date(pub_date)
+                    else:
+                        title = str(a)
+                        source = pub_date = url_item = ""
+                    if self.context.has_rich:
+                        self.context.console.print(f"  [bold]{idx}.[/bold] {title}")
+                        meta_parts = [p for p in [source, pub_date] if p]
+                        if meta_parts:
+                            self.context.console.print(f"     [dim]{' · '.join(meta_parts)}[/dim]")
+                    else:
+                        meta = f" ({source})" if source else ""
+                        print(f"  {idx}. {title}{meta}")
+                if self.context.has_rich:
+                    self.context.console.print()
+            else:
+                # Empty articles — show helpful config guidance
+                _data_keys = _load_data_keys()
+                if self.context.has_rich:
+                    self.context.console.print()
+                    self.context.console.print(f"  [dim]未找到 '{topic}' 的相关新闻。[/dim]")
+                    if not _data_keys.get("finnhub") and not _data_keys.get("newsapi"):
+                        self.context.console.print("  [dim]配置数据服务 key 可获取更多新闻来源：[/dim]")
+                        self.context.console.print("  [dim]  /apikey set finnhub <key>   →  https://finnhub.io/register[/dim]")
+                        self.context.console.print("  [dim]  /apikey set newsapi <key>   →  https://newsapi.org/register[/dim]")
+                    self.context.console.print()
+        else:
+            articles = await asyncio.get_event_loop().run_in_executor(
+                None, _fetch_public_news_fallback, topic, limit
+            )
+            if isinstance(articles, list) and articles:
+                if self.context.has_rich:
+                    self.context.console.print()
+                    self.context.console.print("  [dim]新闻 API 不可用，已使用公共 RSS fallback。[/dim]")
+                    self.context.console.print()
+                for idx, a in enumerate(articles[:limit], 1):
+                    title = a.get("title", "Untitled") if isinstance(a, dict) else str(a)
+                    source = a.get("source", "") if isinstance(a, dict) else ""
+                    pub_date = a.get("published_at", "") if isinstance(a, dict) else ""
+                    if pub_date:
+                        pub_date = _news_date(pub_date)
+                    if self.context.has_rich:
+                        self.context.console.print(f"  [bold]{idx}.[/bold] {title}")
+                        meta_parts = [p for p in [source, pub_date] if p]
+                        if meta_parts:
+                            self.context.console.print(f"     [dim]{' · '.join(meta_parts)}[/dim]")
+                    else:
+                        meta = f" ({source})" if source else ""
+                        print(f"  {idx}. {title}{meta}")
+                if self.context.has_rich:
+                    self.context.console.print()
+                return
+
+            # Backend + all local fallbacks unavailable — show actionable config guide
+            err = result.get("error", "")
+            _data_keys = _load_data_keys()
+            _has_finnhub = bool(_data_keys.get("finnhub"))
+            _has_newsapi = bool(_data_keys.get("newsapi"))
+            if self.context.has_rich:
+                self.context.console.print()
+                self.context.console.print("  [yellow]⚠ 新闻服务不可用[/yellow]")
+                if not _has_finnhub and not _has_newsapi:
+                    self.context.console.print("  [dim]配置以下任意一个数据服务 key 即可获取新闻：[/dim]")
+                    self.context.console.print("  [dim]  Finnhub  (免费60次/分) → /apikey set finnhub <key>   注册: https://finnhub.io/register[/dim]")
+                    self.context.console.print("  [dim]  NewsAPI  (免费100次/天) → /apikey set newsapi <key>   注册: https://newsapi.org/register[/dim]")
+                else:
+                    self.context.console.print(f"  [dim]错误: {err[:120] if err else '获取失败'}[/dim]")
+                self.context.console.print(f"  [dim]或使用: /web {topic} latest news — 通过 Brave 搜索[/dim]")
+                self.context.console.print()
+            else:
+                print("  News unavailable. Configure: /apikey set finnhub <key>")
+
+    async def cmd_quote(self, args: str):
+        symbols = parse_symbols(args, self.terminal.config.get("watchlist", ["AAPL"]))
+
+        # 优先使用 MarketDataClient（真实实时数据，代理绕过）
+        if _get__HAS_MDC():
+            mdc = _get_mdc()
+            if self.context.has_rich:
+                self.context.console.print()
+            for symbol in symbols:
+                if self.context.has_rich:
+                    with self.context.console.status(f"[dim]{symbol}...[/dim]", spinner="dots"):
+                        loop = asyncio.get_event_loop()
+                        r = await loop.run_in_executor(None, mdc.quote, symbol)
+                else:
+                    r = mdc.quote(symbol)
+
+                if r.get("success"):
+                    name    = r.get("name", symbol)
+                    # Supplement Chinese name for A-shares where yfinance returns ASCII
+                    if _is_ashare_symbol(symbol) and (not name or name == symbol or name.replace(" ","").isascii()):
+                        _cn = _ashare_code_to_name(symbol)
+                        if _cn:
+                            name = _cn
+                    print_quote_result(console=self.context.console, has_rich=self.context.has_rich, symbol=symbol, quote=r, name=name)
+                else:
+                    print_quote_result(console=self.context.console, has_rich=self.context.has_rich, symbol=symbol, quote=r)
+            if self.context.has_rich:
+                self.context.console.print()
+            return
+
+        # Fallback：原有 Aria 工具
+        for symbol in symbols:
+            if self.context.has_rich:
+                with self.context.console.status(f"[dim]Fetching {symbol}...[/dim]", spinner="dots"):
+                    result = await execute_aria_tool(self.terminal.api_url, "get_market_data", {
+                        "symbol": symbol, "market": "US", "period": "1mo"
+                    })
+            else:
+                print(f"Fetching {symbol}...")
+                result = await execute_aria_tool(self.terminal.api_url, "get_market_data", {
+                    "symbol": symbol, "market": "US", "period": "1mo"
+                })
+            if not result:
+                print_error(self.context, f"{symbol}: 数据服务不可用（API未运行）", "tool")
+                continue
+            if result.get("success") and result.get("data"):
+                output = format_quote_output(result)
+                self.context.console.print(output)
+            else:
+                print_error(self.context, f"Failed: {result.get('error', 'No data')}")
+
+    async def cmd_screen_cn(self, args: str):
+        """A股选股筛选器 (local, akshare)."""
+        params: Dict[str, Any] = {}
+        for tok in args.split():
+            if "=" in tok:
+                k, v = tok.split("=", 1)
+                params[k.strip()] = v.strip()
+        tool_name = "screen_ashare"
+        if tool_name in _get_LOCAL_TOOLS():
+            await self._run_local_tool(tool_name, params, "A股选股筛选")
+        else:
+            await self.terminal.send_message(f"帮我筛选A股股票，条件：{args or '市值>50亿，非ST，流动性好'}")
+
+    async def cmd_limitup(self, args: str):
+        """A股涨停板池.  Usage: /limitup [YYYY-MM-DD] [code_filter]"""
+        import re as _re_lu
+        arg = args.strip()
+
+        # Detect if arg looks like a stock code (6 digits) vs a date
+        _is_code  = bool(_re_lu.match(r'^[036]\d{5}$', arg))
+        _is_date  = bool(_re_lu.match(r'^\d{4}-\d{2}-\d{2}$', arg))
+        _code_filter = arg if _is_code else None
+        _date_arg    = arg if _is_date else ""
+        params = {"date": _date_arg} if _date_arg else {}
+
+        tool_name = "get_limit_up_pool"
+        if tool_name in _get_LOCAL_TOOLS():
+            await self._run_local_tool(tool_name, params, "涨停板池")
+        else:
+            # Direct akshare fallback — avoids "A股" keyword triggering market snapshot routing
+            try:
+                import akshare as ak
+                from datetime import date as _dt
+                _date_str = (_date_arg.replace("-", "") if _date_arg
+                             else _dt.today().strftime("%Y%m%d"))
+                _df = ak.stock_zt_pool_em(date=_date_str)
+                if _df is not None and not _df.empty:
+                    if _code_filter:
+                        _col = next((c for c in _df.columns if "代码" in str(c) or c == "code"), None)
+                        if _col:
+                            _df = _df[_df[_col].astype(str) == _code_filter]
+                    _count = len(_df)
+                    if self.context.has_rich:
+                        from rich.table import Table
+                        _date_label = _date_arg or _dt.today().isoformat()
+                        tbl = Table(title=f"涨停板池 · {_date_label} · {_count}只", show_header=True, header_style="bold")
+                        _col_map = {"代码": "代码", "名称": "名称", "涨停统计": "涨停统计",
+                                    "连续涨停": "连板", "首次封板时间": "首封", "涨停类型": "类型"}
+                        _show_cols = [c for c in _df.columns if c in _col_map]
+                        for c in _show_cols:
+                            tbl.add_column(_col_map.get(c, c), no_wrap=True)
+                        for _, row in _df.head(30).iterrows():
+                            tbl.add_row(*[str(row[c]) for c in _show_cols])
+                        self.context.console.print(tbl)
+                    else:
+                        print(f"涨停板池 {_count}只")
+                        for _, row in _df.head(20).iterrows():
+                            print(f"  {row.get('代码','')} {row.get('名称','')}")
+                    return
+            except Exception as _e:
+                pass
+            if self.context.has_rich:
+                self.context.console.print("[yellow]akshare 暂不可用，涨停板池无法获取[/yellow]")
+            else:
+                print("akshare unavailable, cannot fetch limit-up pool")
+
+    async def cmd_north(self, args: str):
+        """北向资金净流入."""
+        params = {"days": int(args.strip())} if args.strip().isdigit() else {"days": 10}
+        tool_name = "get_northbound_flow"
+        if tool_name in _get_LOCAL_TOOLS():
+            await self._run_local_tool(tool_name, params, "北向资金")
+        else:
+            await self.terminal.send_message("查询最近10天北向资金（沪深港通）净买入情况")