MonteWalk

Instructions

Implementation Reference

• tools/market_data.py:27-86 (handler)

  Core handler implementation of the 'get_price' MCP tool. Fetches historical stock/crypto price data (OHLCV) from yfinance with retry logic, optional candlestick visualization, and JSON output.
  def get_price( symbol: str, interval: Literal["1m", "2m", "5m", "15m", "30m", "60m", "90m", "1h", "1d", "5d", "1wk", "1mo", "3mo"] = "1d", period: Literal["1d", "5d", "1mo", "3mo", "6mo", "1y", "2y", "5y", "10y", "ytd", "max"] = "1y", visualize: bool = False ) -> str: """ Retrieves historical price data (OHLCV) for a given symbol. Args: symbol: The ticker symbol (e.g., 'AAPL', 'BTC-USD'). interval: Data interval. Valid values: "1m", "2m", "5m", "15m", "30m", "60m", "90m", "1h", "1d", "5d", "1wk", "1mo", "3mo". period: Data period to download. Valid values: "1d", "5d", "1mo", "3mo", "6mo", "1y", "2y", "5y", "10y", "ytd", "max". visualize: If True, returns a candlestick chart instead of raw data. Returns: JSON string of OHLCV data or base64-encoded candlestick chart. """ try: ticker = yf.Ticker(symbol) history = ticker.history(period=period, interval=interval) if history.empty: logger.warning(f"No price data found for {symbol} (period={period}, interval={interval})") return json.dumps([]) # Reset index to make Date a column history.reset_index(inplace=True) if visualize: try: from tools.visualizer import plot_candlestick # Convert to DataFrame format expected by visualizer df = history[['Date', 'Open', 'High', 'Low', 'Close']].copy() if 'Volume' in history.columns: df['Volume'] = history['Volume'] chart = plot_candlestick( df, title=f"{symbol} - {period} ({interval} interval)", volume='Volume' in df.columns ) return chart except Exception as e: logger.error(f"Error generating candlestick chart: {e}") # Fall through to return data # Convert to JSON-friendly format (list of dicts) # Convert timestamps to string history['Date'] = history['Date'].astype(str) data = history.to_dict(orient="records") logger.info(f"Fetched {len(data)} price records for {symbol}") return json.dumps(data, indent=2) except Exception as e: logger.error(f"Error fetching price data for {symbol}: {e}") return json.dumps([])
• server.py:370-373 (registration)

  MCP tool registration of 'get_price' as part of the 'Market Data' category. The register_tools helper applies the @mcp.tool() decorator to enable the tool in the FastMCP server.
  register_tools( [get_price, get_fundamentals, get_orderbook], "Market Data" )
• tools/portfolio_optimizer.py:17-38 (helper)

  Internal helper function named get_price used within portfolio_optimizer.py for fetching price data in optimization workflows. Not the registered MCP tool.
  def get_price(ticker: str, period: str, interval: str) -> str: """ Fetches historical price data for a given ticker and returns it as a JSON string. This is a placeholder to satisfy the new mean_variance_optimize function's dependency. In a real scenario, this might be an API call or a more sophisticated data fetcher. """ try: data = yf.download(ticker, period=period, interval=interval, progress=False) if data.empty: return json.dumps([]) # Rename columns to match expected structure if necessary, e.g., 'Date' instead of index data_reset = data.reset_index() data_reset['Date'] = data_reset['Date'].dt.strftime('%Y-%m-%d') # Format date as string # Select relevant columns, e.g., 'Date', 'Close' # The new code expects 'Date' and 'Close' to be present. # If other columns are needed, adjust here. history_list = data_reset[['Date', 'Close']].to_dict(orient='records') return json.dumps(history_list) except Exception as e: logger.error(f"Error fetching data for {ticker} with yfinance: {e}") return json.dumps([])
• app.py:286-286 (registration)

  Inclusion of get_price in the Gradio UI toolbox under 'Market Data' category for demo interface.
  "Market Data": [get_price, get_fundamentals, get_orderbook],