Tool Calling Fundamentals

import google.generativeai as genai
import json

genai.configure(api_key="YOUR_API_KEY")

# Step 1: Define the tools - name, description, and parameter schema
tools = [
    {
        "name": "get_stock_price",
        "description": "Fetches the current stock price for a given ticker symbol. Call this whenever the user asks about stock prices, market values, or share prices.",
        "parameters": {
            "type": "object",
            "properties": {
                "ticker": {
                    "type": "string",
                    "description": "Stock ticker symbol, e.g. 'AAPL' for Apple, 'GOOGL' for Google"
                }
            },
            "required": ["ticker"]
        }
    }
]

# Step 2: Simulate the actual function your code would run
def get_stock_price(ticker: str) -> dict:
    # In production this calls a real API like Yahoo Finance or Alpha Vantage
    prices = {"AAPL": 189.43, "GOOGL": 175.20, "MSFT": 412.78}
    price = prices.get(ticker.upper(), None)
    if price is None:
        return {"error": f"Ticker {ticker} not found"}
    return {"ticker": ticker.upper(), "price": price, "currency": "USD"}

# Step 3: Send the user message - model decides to use a tool
model = genai.GenerativeModel("gemini-1.5-flash", tools=tools)
response = model.generate_content("What is Apple's stock price right now?")

# Step 4: Model returns a function call, not plain text
part = response.candidates[0].content.parts[0]
if hasattr(part, "function_call"):
    fn = part.function_call
    print(f"Model wants to call: {fn.name}({dict(fn.args)})")
    # → Model wants to call: get_stock_price({'ticker': 'AAPL'})
    
    # Step 5: Execute the real function
    result = get_stock_price(**dict(fn.args))
    print(f"Function result: {result}")
    # → Function result: {'ticker': 'AAPL', 'price': 189.43, 'currency': 'USD'}
    
    # Step 6: Send result back - model synthesises the final answer
    # (conversation continuation with function response)