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 xml.etree import ElementTree

@tool
def get_lastest_filing_info(company: str = None, cik: str = None, type: str = "10-K", dateb: str = None, start: str = '0', count: str = '50') -> str:
    """A tool that gets the most recent 10-K filings (including addendums) of a company from the SEC.
    Args:
        company: The name of the company.
        cik: The Central Index Key (CIK) of the company.
        type: The type of filing. (Default is "10-K" for this function.)
        dateb: The date before which the filings were made (format: YYYYMMDD).
        start: The starting point for pagination.
        count: The number of filings to retrieve per page.
    Returns:
        The latest filings of a company from the SEC.
    """
    # Validate dateb format
    if dateb:
        try:
            datetime.datetime.strptime(dateb, "%Y%m%d")
        except ValueError as e:
            print(f"Date validation error: {e}")
            return {"error": str(e)}
    
    # Ensure dateb is not after the current time
    current_date = datetime.datetime.now().strftime("%Y%m%d")
    if dateb and dateb > current_date:
        return {"error": "dateb cannot be after the current date."}
    
    base_url = "https://www.sec.gov/cgi-bin/browse-edgar?action=getcurrent"
    params = {
        "CIK": cik,
        "type": type,
        "company": company,
        "dateb": dateb,
        "owner": "include",
        "start": start,
        "count": count,
        "output": "atom"
    }
    # Set a proper User-Agent header per SEC guidelines
    headers = {
        "User-Agent": "FuncPhenomenon SMOLAgentTest/1.0"
    }
    
    response = requests.get(base_url, params=params, headers=headers)
    if response.status_code != 200:
        print(f"SEC retrieval error: {response.text}")
        return {"error": response.text}
    
    filings = []
    try:
        feed = response.content
        root = ElementTree.fromstring(feed)
        # Process filings from the first page
        for entry in root.findall("{http://www.w3.org/2005/Atom}entry"):
            filing = {
                "title": entry.find("{http://www.w3.org/2005/Atom}title").text,
                "link": entry.find("{http://www.w3.org/2005/Atom}link").attrib['href'],
                "summary": entry.find("{http://www.w3.org/2005/Atom}summary").text,
                "updated": entry.find("{http://www.w3.org/2005/Atom}updated").text,
                "category": entry.find("{http://www.w3.org/2005/Atom}category").attrib['term'],
                "id": entry.find("{http://www.w3.org/2005/Atom}id").text
            }
            # Only include filings where category starts with "10-K"
            if filing["category"].startswith("10-K"):
                filings.append(filing)
        
        # Fetch additional pages until we have 100 filings or no more are available
        while len(filings) < 100:
            params["start"] = str(int(params["start"]) + int(params["count"]))
            response = requests.get(base_url, params=params, headers=headers)
            if response.status_code != 200:
                print(f"SEC retrieval error on pagination: {response.text}")
                break
            feed = response.content
            root = ElementTree.fromstring(feed)
            entries = root.findall("{http://www.w3.org/2005/Atom}entry")
            if not entries:
                break
            for entry in entries:
                filing = {
                    "title": entry.find("{http://www.w3.org/2005/Atom}title").text,
                    "link": entry.find("{http://www.w3.org/2005/Atom}link").attrib['href'],
                    "summary": entry.find("{http://www.w3.org/2005/Atom}summary").text,
                    "updated": entry.find("{http://www.w3.org/2005/Atom}updated").text,
                    "category": entry.find("{http://www.w3.org/2005/Atom}category").attrib['term'],
                    "id": entry.find("{http://www.w3.org/2005/Atom}id").text
                }
                if filing["category"].startswith("10-K"):
                    filings.append(filing)
                    if len(filings) >= 100:
                        break

    except Exception as e:
        return {"error": f"Failed to parse XML: {str(e)}"}
    
    return filings

@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:
        # Create timezone object
        tz = pytz.timezone(timezone)
        # Get current time in that 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)}"


final_answer = FinalAnswerTool()

# If the agent does not answer, the model is overloaded, please use another model or the following Hugging Face Endpoint that also contains qwen2.5 coder:
# model_id='https://pflgm2locj2t89co.us-east-1.aws.endpoints.huggingface.cloud' 

model = HfApiModel(
max_tokens=2096,
temperature=0.5,
model_id='Qwen/Qwen2.5-Coder-32B-Instruct',# it is possible that this model may be overloaded
custom_role_conversions=None,
)


# Import tool from Hub
image_generation_tool = load_tool("agents-course/text-to-image", trust_remote_code=True)

with open("prompts.yaml", 'r') as stream:
    prompt_templates = yaml.safe_load(stream)
    
agent = CodeAgent(
    model=model,
    tools=[get_lastest_filing_info, get_current_time_in_timezone, final_answer], ## add your tools here (don't remove final answer)
    max_steps=6,
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name=None,
    description=None,
    prompt_templates=prompt_templates
)


GradioUI(agent).launch()