Update app.py

app.py

@@ -20,35 +20,38 @@ llm = HuggingFaceEndpoint(
 # Streamlit UI setup
 st.set_page_config(page_title="🧠 HuggingFace Chatbot", page_icon="🤖")
 st.title("🤖 HuggingFace Chatbot")
-st.caption("Built with Streamlit + LangChain (no schema, limited to 50 words per answer)")
+st.caption("Built with Streamlit + LangChain (50-word max answers)")
 
-# Initialize session state for chat history
+# Initialize chat history
 if "messages" not in st.session_state:
     st.session_state.messages = [
         {"role": "assistant", "content": "Hi there! Ask me anything."}
     ]
 
-# Display chat history
+# Display chat messages
 for msg in st.session_state.messages:
     with st.chat_message(msg["role"]):
         st.markdown(msg["content"])
 
-# Handle user input
+# Chat input
 if prompt := st.chat_input("Type your message here..."):
+    # Add user message
     st.session_state.messages.append({"role": "user", "content": prompt})
     with st.chat_message("user"):
         st.markdown(prompt)
 
-    # Build the conversation history with a system-like instruction
-    instruction = "You are a helpful and concise assistant. Always respond in 50 words or fewer.\n\n"
-    chat_history = ""
+    # Construct prompt (only user + assistant, formatted)
+    conversation = "You are a helpful assistant. Keep replies within 50 words.\n\n"
     for msg in st.session_state.messages:
-        role = "User" if msg["role"] == "user" else "Assistant"
-        chat_history += f"{role}: {msg['content']}\n"
-    context = instruction + chat_history + "Assistant:"
+        if msg["role"] == "user":
+            conversation += f"User: {msg['content']}\n"
+        elif msg["role"] == "assistant":
+            continue  # Don't include previous assistant replies
 
-    # Generate response
+    conversation += "Assistant:"  # Prompt the model to continue
+
+    # Generate model response
     with st.chat_message("assistant"):
-        response = llm.invoke(context)
+        response = llm.invoke(conversation).strip().split("Assistant:")[-1].strip()
         st.markdown(response)
         st.session_state.messages.append({"role": "assistant", "content": response})