youtube_tools.py

from langchain.tools import Tool
from youtube_transcript_api import YouTubeTranscriptApi, NoTranscriptFound, TranscriptsDisabled

def extract_youtube_transcript(youtube_url: str) -> str:
    """
    Extracts the transcript from a given YouTube video URL.
    Returns the transcript as a single string or an error message if not found.
    """
    try:
        video_id = youtube_url.split("v=")[1].split("&")[0]
        transcript_list = YouTubeTranscriptApi.get_transcript(video_id)
        transcript = " ".join([item['text'] for item in transcript_list])
        return transcript
    except NoTranscriptFound:
        return "Error: No transcript found for this video. It might be disabled or not available in English."
    except TranscriptsDisabled:
        return "Error: Transcripts are disabled for this video."
    except Exception as e:
        return f"Error extracting transcript: {str(e)}"

youtube_transcript_tool = Tool(
    name="youtube_transcript_extractor",
    func=extract_youtube_transcript,
    description="Extracts the full transcript from a YouTube video given its URL. Input should be a valid YouTube video URL."
)