removed step 8 as it is not working because of it

app.py

@@ -115,34 +115,34 @@ class ChatHistory:
 chat_history = ChatHistory()
 
 # Step 5: Self-Verification and Content Moderation
-def verify_health_wellness_query(query, retrieved_data):
-    """
-    Verifies if the query is related to health and wellness and checks if retrieved data is relevant.
-    """
-    query_lower = query.lower()
-
-    # Use Groq's LLM to evaluate the safety of the query (new LLM-based moderation)
-    chat_completion = client.chat.completions.create(
-        messages=[
-            {"role": "user", "content": query}
-        ],
-        model="llama-guard-3-8b",
-    )
+# def verify_health_wellness_query(query, retrieved_data):
+#     """
+#     Verifies if the query is related to health and wellness and checks if retrieved data is relevant.
+#     """
+#     query_lower = query.lower()
+
+#     # Use Groq's LLM to evaluate the safety of the query (new LLM-based moderation)
+#     chat_completion = client.chat.completions.create(
+#         messages=[
+#             {"role": "user", "content": query}
+#         ],
+#         model="llama-guard-3-8b",
+#     )
     
-    moderation_result = chat_completion.choices[0].message.content.strip()
+#     moderation_result = chat_completion.choices[0].message.content.strip()
 
-    # If the model's response indicates harmful content, block the query
-    if 'unsafe' in moderation_result.lower() or 'harmful' in moderation_result.lower():
-        return False, "The query is flagged as unsafe or harmful. Please rephrase it."
+#     # If the model's response indicates harmful content, block the query
+#     if 'unsafe' in moderation_result.lower() or 'harmful' in moderation_result.lower():
+#         return False, "The query is flagged as unsafe or harmful. Please rephrase it."
 
-    # Proceed with verifying if the retrieved data aligns with the health/wellness context
-    wellness_keywords = ['mental health', 'stress', 'wellness', 'anxiety', 'relaxation', 'meditation']
+#     # Proceed with verifying if the retrieved data aligns with the health/wellness context
+#     wellness_keywords = ['mental health', 'stress', 'wellness', 'anxiety', 'relaxation', 'meditation']
 
-    # Check if the query contains any of the relevant wellness-related keywords
-    if any(keyword in query_lower for keyword in wellness_keywords):
-        return True, ""  # If any relevant wellness keyword is present, it's valid
+#     # Check if the query contains any of the relevant wellness-related keywords
+#     if any(keyword in query_lower for keyword in wellness_keywords):
+#         return True, ""  # If any relevant wellness keyword is present, it's valid
 
-    return False, "The query does not seem to match health and wellness topics."
+#     return False, "The query does not seem to match health and wellness topics."
 
 # Step 6: Define Gradio Interface for Chatbot
 def gradio_chatbot(user_query):
@@ -150,8 +150,8 @@ def gradio_chatbot(user_query):
     retrieved_data = chatbot_response(user_query)
 
     # Step 8: Check and verify the query for health/wellness content
-    is_valid, message = verify_health_wellness_query(user_query, retrieved_data)
-    
+    # is_valid, message = verify_health_wellness_query(user_query, retrieved_data)
+    is_valid = True
     if is_valid:
         # Generate a coherent response using Groq's DeepSeek-R1 LLM
         coherent_response = generate_coherent_response(user_query, retrieved_data, chat_history.get_history())