tools.py

from langchain_core.tools import tool
from langchain_community.tools.tavily_search import TavilySearchResults
from youtube_transcript_api import YouTubeTranscriptApi

@tool
def add(a: int, b: int) -> int:
    """
    Sums two values and returns the result of the sum

    Args:
        a: first number
        b: second number
    """
    return a + b

@tool
def substract(a: int, b: int) -> int:
    """
    Subtracts one value from another and returns the result of the sum

    Args:
        a: first number
        b: second number
    """
    return a - b

@tool
def multiply(a: int, b: int) -> int:
    """
    Multiplies two values and returns the result of the sum

    Args:
        a: first number
        b: second number
    """
    return a * b

@tool
def divide(a: int, b: int) -> int:
    """
    Divides two values and returns the result of the sum

    Args:
        a: numerator
        b: denominator
    """
    if b == 0:
        raise ValueError("Cannot divide by zero.")
    return a / b

@tool
def web_search(query: str) -> str:
    """Search Tavily for a query and return maximum 3 results.
    
    Args:
        query: The search query."""
    search_docs = TavilySearchResults(max_results=3).invoke(query=query)
    formatted_search_docs = "\n\n---\n\n".join(
        [
            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
            for doc in search_docs
        ])
    return {"web_results": formatted_search_docs}

@tool
def fetch_youtube_transcript(video_url: str) -> str:
    """
    Fetch transcript of a YouTube video by URL.
    Returns plain text if transcript is available.
    """
    import re
    video_id_match = re.search(r"(?:v=|youtu\.be/)([\w-]+)", video_url)
    if not video_id_match:
        return "Invalid YouTube URL."

    video_id = video_id_match.group(1)

    try:
        transcript_list = YouTubeTranscriptApi.get_transcript(video_id, languages=['en'])
        transcript_text = " ".join([entry['text'] for entry in transcript_list])
        return transcript_text[:4000]  # Limit to fit into context window
    except Exception as e:
        return f"Transcript not available or error: {str(e)}"