Yahoo Finance MCP Server

Instructions

Implementation Reference

• server.py:68-124 (handler)

  The async handler function that implements the tool logic: fetches data from Financial Modeling Prep API based on interval, processes with pandas DataFrame, filters by period, and returns JSON.

  async def get_historical_stock_prices(
      ticker: str, period: str = "1mo", interval: str = "1d"
  ) -> str:
      """Get historical stock prices for a given ticker symbol"""
  
      api_key = os.environ.get("FMP_API_KEY")
      if not api_key:
          return "Error: FMP_API_KEY environment variable not set."
  
      base = "https://financialmodelingprep.com/api/v3"
      try:
          if interval in ["1min", "5min", "15min", "30min", "1hour", "4hour"]:
              url = f"{base}/historical-chart/{interval}/{ticker}"
              resp = requests.get(url, params={"apikey": api_key}, timeout=10)
              resp.raise_for_status()
              data = resp.json()
          else:
              url = f"{base}/historical-price-full/{ticker}"
              resp = requests.get(url, params={"apikey": api_key}, timeout=10)
              resp.raise_for_status()
              data = resp.json().get("historical", [])
      except Exception as e:
          return f"Error: getting historical stock prices for {ticker}: {e}"
  
      df = pd.DataFrame(data)
      if not df.empty:
          if "date" in df.columns:
              df.rename(columns={"date": "Date"}, inplace=True)
          df = df[["Date", "open", "high", "low", "close", "volume"]]
          df.columns = [c.title() for c in df.columns]
          # Convert the Date column to datetime for proper comparison
          df["Date"] = pd.to_datetime(df["Date"])
  
      period_map = {
          "1d": pd.Timedelta(days=1),
          "5d": pd.Timedelta(days=5),
          "1mo": pd.Timedelta(days=30),
          "3mo": pd.Timedelta(days=90),
          "6mo": pd.Timedelta(days=180),
          "1y": pd.Timedelta(days=365),
          "2y": pd.Timedelta(days=730),
          "5y": pd.Timedelta(days=1825),
          "10y": pd.Timedelta(days=3650),
      }
  
      if period == "ytd":
          start = pd.Timestamp.now().replace(month=1, day=1)
      elif period != "max" and period in period_map:
          start = pd.Timestamp.now() - period_map[period]
      else:
          start = None
  
      if start is not None and not df.empty:
          df = df[df["Date"] >= start]
  
      return df.to_json(orient="records", date_format="iso")

• server.py:54-67 (registration)

  Tool registration decorator @fmp_server.tool with name and detailed description including input parameters.

  @fmp_server.tool(
      name="get_historical_stock_prices",
      description="""获取指定股票的历史价格，返回日期、开盘价、最高价、最低价、收盘价和成交量。
  参数说明：
      ticker: str
          股票代码，例如 "AAPL"
      period : str
          支持的周期：1d,5d,1mo,3mo,6mo,1y,2y,5y,10y,ytd,max；
          也可以使用开始和结束日期；默认 "1mo"
      interval : str
          支持的间隔：1m,2m,5m,15m,30m,60m,90m,1h,1d,5d,1wk,1mo,3mo；
          分钟级数据最多可追溯60天；默认 "1d"
  """,
  )

• server.py:68-70 (schema)

  Function signature providing type hints for input parameters (ticker, period, interval) and return type (str).

  async def get_historical_stock_prices(
      ticker: str, period: str = "1mo", interval: str = "1d"
  ) -> str: