Update app.py

app.py

@@ -5,6 +5,7 @@ import inspect
 import pandas as pd
 from smolagents import CodeAgent, InferenceClientModel, DuckDuckGoSearchTool, tool
 from smolagents import OpenAIServerModel
+from youtube_transcript_api import YouTubeTranscriptApi
 
 
 # (Keep Constants as is)
@@ -51,7 +52,23 @@ def analyze_image(question: str, image_url: str) -> str:
 
     return response
 
-
+@tool
+# a function to get youtube transcript from video id
+def get_youtube_transcript(video_id: str) -> str:
+    """
+    Fetches the transcript of a YouTube video given its video ID.    
+    Args:
+        video_id (str): The ID of the YouTube video. Pass in the video ID, NOT the video URL. For a video with the URL https://www.youtube.com/watch?v=12345 the ID is 12345.
+    Returns:
+        str: The transcript of the YouTube video. as a single string with each line separated by a newline character.
+    """
+    # Initialize the YouTubeTranscriptApi
+    ytt_api = YouTubeTranscriptApi()
+    fetched_transcript = ytt_api.fetch(video_id)
+    raw_data = fetched_transcript.to_raw_data()
+    # raw data is in the form of [{        'text': 'Hey there',        'start': 0.0,        'duration': 1.54    },    {        'text': 'how are you',,        'start': 1.54,        'duration': 4.16    },   ... ] we will return ony the text element as lines
+    transcript = "\n".join([item['text'] for item in raw_data])
+    return transcript
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
@@ -83,7 +100,7 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         search_tool = DuckDuckGoSearchTool()
         my_small_agent = CodeAgent(
                         model=openai_41mini_model,
-                        tools=[search_tool, analyze_image],
+                        tools=[search_tool, analyze_image, get_youtube_transcript],
                         name="web_agent",
                         description="Use search engine to find webpages related to a subject and get the page content",
                         #additional_authorized_imports=["pandas", "numpy","bs4"],