app.py

import datetime
import pytz
import yaml
import yfinance as yf

from smolagents import CodeAgent, HfApiModel, load_tool, tool
from tools.final_answer import FinalAnswerTool
from Gradio_UI import GradioUI

# Tool to fetch the current stock price using yfinance.
@tool
def get_current_stock_price(symbol: str) -> str:
    """Fetches the current stock price for the given stock symbol.
    
    Args:
        symbol: The stock ticker symbol (e.g., 'AAPL').
        
    Returns:
        A string reporting the current stock price.
    """
    try:
        ticker = yf.Ticker(symbol)
        price = ticker.info.get("regularMarketPrice", None)
        if price is None:
            return f"Could not fetch current price for {symbol}."
        return f"The current price of {symbol} is {price:.2f} USD."
    except Exception as e:
        return f"Error fetching current price for {symbol}: {str(e)}"

# Tool to fetch historical stock data for a given timeline using yfinance.
@tool
def get_stock_historical_data(symbol: str, timeline: str) -> str:
    """Fetches historical stock data for a given symbol and timeline.
    
    Args:
        symbol: The stock ticker symbol (e.g., 'TSLA').
        timeline: The timeframe for data. Acceptable values: '1h', '1day', '1week', '1year'.
        
    Returns:
        A summary string of the historical price data.
    """
    try:
        ticker = yf.Ticker(symbol)
        if timeline == "1h":
            # For 1h data, fetch 1 day of data at 1-minute intervals and take the last 60 minutes.
            df = ticker.history(period="1d", interval="1m").tail(60)
        elif timeline == "1day":
            df = ticker.history(period="1d", interval="5m")
        elif timeline == "1week":
            df = ticker.history(period="7d", interval="30m")
        elif timeline == "1year":
            df = ticker.history(period="1y", interval="1d")
        else:
            return f"Timeline '{timeline}' is not supported. Use '1h', '1day', '1week', or '1year'."
        
        if df.empty:
            return f"No historical data available for {symbol} over timeline {timeline}."
        
        start_price = df['Close'].iloc[0]
        end_price = df['Close'].iloc[-1]
        min_price = df['Close'].min()
        max_price = df['Close'].max()
        summary = (
            f"For {symbol} over {timeline}:\n"
            f"Start Price: {start_price:.2f} USD\n"
            f"End Price: {end_price:.2f} USD\n"
            f"Min Price: {min_price:.2f} USD\n"
            f"Max Price: {max_price:.2f} USD"
        )
        return summary
    except Exception as e:
        return f"Error fetching historical data for {symbol} over {timeline}: {str(e)}"

# Final answer tool (must be included)
final_answer = FinalAnswerTool()

# Define the model configuration.
# Here we reduce max_tokens to help keep the total token count below the limit.
model = HfApiModel(
    max_tokens=1000,  # Reduced from 2096 to reduce total token count
    temperature=0.5,
    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
    custom_role_conversions=None,
)

# Optionally, load an additional tool from the Hub (e.g., a text-to-image tool)
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)

# Load prompt templates from YAML.
with open("prompts.yaml", 'r') as stream:
    prompt_templates = yaml.safe_load(stream)
    
# Initialize the CodeAgent with the new yfinance tools.
agent = CodeAgent(
    model=model,
    tools=[final_answer, get_current_stock_price, get_stock_historical_data],
    max_steps=3,  # Reduced steps to limit the chain-of-thought length
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name="TradingAgent",
    description="An AI agent that analyzes stock data using yfinance and advises on long-term trading decisions.",
    prompt_templates=prompt_templates
)

# Launch the Gradio UI to interact with the agent.
GradioUI(agent).launch()