Update app.py

app.py

@@ -15,6 +15,40 @@ openai_key = os.environ.get("OPENAI_API_KEY")
 
 search_tool = DuckDuckGoSearchTool()
 
+##Tool 1
+import re
+from youtube_transcript_api import YouTubeTranscriptApi
+from smolagents import Tool
+
+class YouTubeCaptionTool(Tool):
+    name = "youtube_caption_reader"
+    description = "Extracts captions from a YouTube video given its URL and returns the transcript or a segment."
+
+    inputs = {
+        "url": {
+            "type": "string",
+            "description": "Full YouTube video URL (e.g., https://www.youtube.com/watch?v=abc123)"
+        }
+    }
+
+    output_type = "string"
+
+    def forward(self, url: str) -> str:
+        try:
+            # Extract the video ID from the URL
+            match = re.search(r"(?:v=|youtu.be/)([\w-]+)", url)
+            if not match:
+                return "Could not extract video ID from URL."
+
+            video_id = match.group(1)
+            transcript = YouTubeTranscriptApi.get_transcript(video_id)
+            full_text = " ".join([entry['text'] for entry in transcript])
+
+            return full_text[:3000]  # return first 3000 characters
+
+        except Exception as e:
+            return f"Failed to retrieve transcript: {str(e)}"
+
 
 ##Tool 2
 import wikipedia
@@ -75,6 +109,7 @@ class WikipediaQATool(Tool):
 
 wiki_tool = WikipediaQATool()
 #excel_tool = ExcelAnalysisTool()
+yt_tool = YouTubeCaptionTool()
 
 async def run_and_submit_all(profile: gr.OAuthProfile | None):
     log_output = ""
@@ -82,7 +117,7 @@ async def run_and_submit_all(profile: gr.OAuthProfile | None):
     try:
         
         agent = ToolCallingAgent(
-            tools=[search_tool],
+            tools=[search_tool, yt_tool],
                         model=OpenAIServerModel(
                     model_id="gpt-4o",            # ✅ valid OpenAI model name
                     temperature=0.0,