app.py

from smolagents import CodeAgent, DuckDuckGoSearchTool, HfApiModel, load_tool, tool
import datetime
import requests
import pytz
import yaml
from tools.final_answer import FinalAnswerTool
from Gradio_UI import GradioUI
from PIL import Image
import io

@tool
def web_researcher(query: str, max_results: int = 3) -> str:
    """A tool that performs web research using DuckDuckGo search.
    Args:
        query: The search query string
        max_results: Maximum number of results to return (default 3)
    Returns:
        A formatted string containing search results
    """
    try:
        search_results = DuckDuckGoSearchTool().search(query, max_results=max_results)
        if not search_results:
            return "No results found for your query."
        
        formatted_results = []
        for i, result in enumerate(search_results, 1):
            formatted_results.append(
                f"Result {i}:\n"
                f"Title: {result.get('title', 'No title')}\n"
                f"URL: {result.get('link', 'No URL')}\n"
                f"Snippet: {result.get('snippet', 'No description')}\n"
                f"{'-'*50}"
            )
        return "\n".join(formatted_results)
    except Exception as e:
        return f"Error performing search: {str(e)}"

@tool
def get_current_time_in_timezone(timezone: str) -> str:
    """A tool that fetches the current local time in a specified timezone.
    Args:
        timezone: A string representing a valid timezone (e.g., 'America/New_York').
    """
    try:
        tz = pytz.timezone(timezone)
        local_time = datetime.datetime.now(tz).strftime("%Y-%m-%d %H:%M:%S")
        return f"The current local time in {timezone} is: {local_time}"
    except Exception as e:
        return f"Error fetching time for timezone '{timezone}': {str(e)}"

# Custom Final Answer Processor
class CustomFinalAnswerTool:
    def __init__(self):
        self.tool = FinalAnswerTool()
    
    def __call__(self, answer):
        # Handle different answer types
        if hasattr(answer, '_image'):  # If it's an image
            return self._process_image_answer(answer)
        elif isinstance(answer, str):  # Regular text answer
            return answer
        else:  # Fallback for other types
            return str(answer)
    
    def _process_image_answer(self, image_answer):
        """Process image answers to display properly"""
        try:
            # Convert the image to displayable format
            img = image_answer._image
            if isinstance(img, Image.Image):
                return img
            elif isinstance(img, bytes):
                return Image.open(io.BytesIO(img))
            return f"Generated image: {image_answer}"
        except Exception as e:
            return f"Image processing error: {str(e)}"

# Initialize tools
final_answer = CustomFinalAnswerTool()
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)

model = HfApiModel(
    max_tokens=2096,
    temperature=0.5,
    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
    custom_role_conversions=None,
)

with open("prompts.yaml", 'r') as stream:
    prompt_templates = yaml.safe_load(stream)
    
agent = CodeAgent(
    model=model,
    tools=[final_answer.tool, web_researcher, get_current_time_in_timezone, image_generation_tool],
    max_steps=10,
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name=None,
    description=None,
    prompt_templates=prompt_templates
)

# Custom UI processor
class CustomGradioUI(GradioUI):
    def process_final_answer(self, final_answer):
        """Override to handle different answer types"""
        if isinstance(final_answer, Image.Image):
            return final_answer
        return super().process_final_answer(final_answer)

CustomGradioUI(agent).launch()