compute_indicators
Calculate technical indicators like RSI, MACD, and BBANDS for financial symbols to support trading strategy analysis and portfolio management.
Instructions
Calculates technical indicators for a symbol.
Args:
symbol: Ticker symbol.
indicators: List of indicators (e.g., ['RSI', 'MACD', 'BBANDS']).Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| symbol | Yes | ||
| indicators | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- tools/feature_engineering.py:9-41 (handler)The primary handler function implementing the compute_indicators tool. Downloads 1 year of historical data for the given symbol using yfinance, computes the specified technical indicators (RSI, MACD, Bollinger Bands) using pandas_ta, handles data structure issues, and returns the last 10 rows of results as JSON string.
def compute_indicators(symbol: str, indicators: List[str] = ["RSI", "MACD"]) -> str: """ Calculates technical indicators for a symbol. Args: symbol: Ticker symbol. indicators: List of indicators (e.g., ['RSI', 'MACD', 'BBANDS']). """ df = yf.download(symbol, period="1y", progress=False) if df.empty: return f"No data for {symbol}" # Handle MultiIndex if isinstance(df.columns, pd.MultiIndex): df.columns = df.columns.get_level_values(0) result = df[['Close']].copy() for ind in indicators: try: if ind.upper() == "RSI": result['RSI'] = ta.rsi(df['Close']) elif ind.upper() == "MACD": macd = ta.macd(df['Close']) result = pd.concat([result, macd], axis=1) elif ind.upper() == "BBANDS": bb = ta.bbands(df['Close']) result = pd.concat([result, bb], axis=1) # Add more as needed except Exception as e: return f"Error computing {ind}: {str(e)}" return result.tail(10).to_json(orient="index") - server.py:390-393 (registration)Registration of the compute_indicators tool (along with related functions) as part of the 'Feature Engineering' category in the MCP server using the register_tools helper function, which applies @mcp.tool() decorator to each.
register_tools( [compute_indicators, rolling_stats, get_technical_summary], "Feature Engineering" ) - server.py:355-364 (helper)The register_tools helper function that dynamically registers imported tools like compute_indicators by applying the MCP @mcp.tool() decorator in a loop with logging.
def register_tools(tools: List[Callable], category: str): """Helper to register multiple tools with logging.""" for tool in tools: try: mcp.tool()(tool) logger.info(f"Registered {category} tool: {tool.__name__}") except Exception as e: logger.error(f"Failed to register {tool.__name__}: {e}") raise - app.py:291-291 (registration)compute_indicators is included in the 'Technical Analysis' tool category for the Gradio UI toolbox (non-MCP usage).
"Technical Analysis": [compute_indicators, rolling_stats, get_technical_summary],