agent_tools.py

from smolagents import tool, Tool
from tavily import TavilyClient
import app_tokens

@tool
def web_search(query: str, proxy: bool = False) -> str:
    """Searches the web for your query.

    Args:
        query: Your query.
        proxy: An optional boolean parameter, if a local proxy should be used or not.

    """
    # tavily_client = TavilyClient(api_key=os.getenv("TAVILY_API_KEY"))
    if proxy:
        tavily_client = TavilyClient(api_key=app_tokens.get_tavily_api_key(), proxies={'http': 'http://localhost:3128', 'https': 'http://localhost:3128'}, verify=False)
    else:
        tavily_client = TavilyClient(api_key=app_tokens.get_tavily_api_key())
    response = tavily_client.search(query)
    return str(response["results"])

class VisitWebpageTool(Tool):
    name = "visit_webpage"
    description = (
        "Visits a webpage at the given url and reads its content as a markdown string. Use this to browse webpages."
    )
    inputs = {
        "url": {
            "type": "string",
            "description": "The url of the webpage to visit.",
        },
        "proxy": {
            "type": "boolean",
            "description": "An optional boolean parameter, if a local proxy should be used or not. Should be True, if the request timed out.",
            "nullable": "True",
        },
    }
    output_type = "string"

    def forward(self, url: str, proxy: bool = True) -> str:
        try:
            import re

            import requests
            from markdownify import markdownify
            from requests.exceptions import RequestException

            from smolagents.utils import truncate_content
        except ImportError as e:
            raise ImportError(
                "You must install packages `markdownify` and `requests` to run this tool: for instance run `pip install markdownify requests`."
            ) from e
        try:
            if proxy:
                response = requests.get(url, timeout=20, proxies={'http': 'http://localhost:3128', 'https': 'http://localhost:3128'}, verify=False)
            else:
                response = requests.get(url, timeout=20)
            response.raise_for_status()  # Raise an exception for bad status codes
            markdown_content = markdownify(response.text).strip()
            markdown_content = re.sub(r"\n{3,}", "\n\n", markdown_content)
            return truncate_content(markdown_content, 40000)

        except requests.exceptions.Timeout:
            return "The request timed out. Please try again later or check the URL."
        except RequestException as e:
            return f"Error fetching the webpage: {str(e)}"
        except Exception as e:
            return f"An unexpected error occurred: {str(e)}"

# test = VisitWebpageTool()
# print(test.forward("https://www.wikipedia.de", proxy=True))