Update app.py

app.py

@@ -20,8 +20,9 @@ TOOL_NAME = "Robot_MCP_Server_robot_watch"
 HTTP_TRANSPORT = StreamableHttpTransport(url=MCP_SERVER_URL)
 MCP_CLIENT = Client(transport=HTTP_TRANSPORT, name=SERVER_NAME)
 
+
 # -------------------------------
-# Async function using user's token
+# Async function using user's HF token
 # -------------------------------
 async def process_webcam_stream_async(image, oauth_token: gr.OAuthToken | None):
     """
@@ -33,7 +34,7 @@ async def process_webcam_stream_async(image, oauth_token: gr.OAuthToken | None):
     if image is None:
         return "", "", "", "", "", "", "", ""
 
-    # Convert image to Base64 string
+    # Convert image to Base64
     buffered = io.BytesIO()
     image.save(buffered, format="JPEG")
     b64_img = base64.b64encode(buffered.getvalue()).decode("utf-8")
@@ -56,6 +57,7 @@ async def process_webcam_stream_async(image, oauth_token: gr.OAuthToken | None):
             response_dict = ast.literal_eval(raw_text)
             vlm_result = response_dict.get("result", {})
 
+            # Extract all fields
             description_out = vlm_result.get("description", "")
             human_out = vlm_result.get("human", "")
             environment_out = vlm_result.get("environment", "")
@@ -65,6 +67,7 @@ async def process_webcam_stream_async(image, oauth_token: gr.OAuthToken | None):
             hazards_list = vlm_result.get("hazards", [])
             objects_list = vlm_result.get("objects", [])
 
+            # Convert lists to strings
             objects_str = ", ".join(objects_list) if isinstance(objects_list, list) else str(objects_list)
             animals_str = ", ".join(animals_list) if isinstance(animals_list, list) else str(animals_list)
             hazards_str = ", ".join(hazards_list) if isinstance(hazards_list, list) else str(hazards_list)
@@ -81,15 +84,17 @@ async def process_webcam_stream_async(image, oauth_token: gr.OAuthToken | None):
             )
 
     except Exception as e:
-        print(f"Error calling remote MCP API: {e}")
+        print(f"Error calling MCP API: {e}")
         import traceback
         traceback.print_exc()
         return f"Error: {e}", "", "", "", "", "", "", ""
 
+
 # -------------------------------
 # Gradio UI
 # -------------------------------
 with gr.Blocks() as demo:
+    # Hugging Face OAuth login button
     gr.LoginButton()
     gr.Markdown("## 🎥 Robot Vision Webcam Stream (MCP Client)")
     with gr.Row():
@@ -104,9 +109,10 @@ with gr.Blocks() as demo:
             objects_out = gr.Textbox(label="Objects Detected", lines=2)
             hazards_out = gr.Textbox(label="Hazards Identified", lines=2)
 
+    # Only webcam input in inputs; Gradio automatically injects oauth_token
     webcam_input.stream(
         process_webcam_stream_async,
-        inputs=[webcam_input, gr.OAuthToken()],
+        inputs=[webcam_input],
         outputs=[
             description_out,
             environment_out,