Update app.py

app.py

@@ -2,10 +2,10 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
-# 🔹 Load HF token from Space Secrets (set in Space settings → Secrets)
+# 🔹 Load HF token from Space Secrets
 HF_TOKEN = os.environ.get('telemedpro')
 
-# 🔹 Default system persona message
+# 🔹 Fixed persona system message
 SYSTEM_MESSAGE = (
     "You are Dr. Alex, a highly knowledgeable yet empathetic doctor. "
     "You always provide clear, safe, and well-structured medical advice in simple language. "
@@ -13,22 +13,18 @@ SYSTEM_MESSAGE = (
     "You behave politely, patiently, and with care, like a trusted family doctor."
 )
 
-# 🔹 Initialize client once
+# 🔹 Initialize InferenceClient once
 client = InferenceClient(token=HF_TOKEN, model="m42-health/Llama3-Med42-70B")
 
-# 🔹 Respond function (non-OAuthToken, stable streaming)
+# 🔹 Respond function
 def respond(message, history, system_message=SYSTEM_MESSAGE, max_tokens=512, temperature=0.7, top_p=0.95):
     try:
         # Start with system message
         messages = [{"role": "system", "content": system_message}]
 
-        # Append previous conversation (Gradio handles history as list of [user, assistant])
+        # Append previous conversation safely
         if history:
-            for h in history:
-                user_msg = h[0] if h[0] else ""
-                ai_msg = h[1] if h[1] else ""
-                messages.append({"role": "user", "content": user_msg})
-                messages.append({"role": "assistant", "content": ai_msg})
+            messages.extend(history)  # ✅ safe, don't manipulate
 
         # Append current user message
         messages.append({"role": "user", "content": message})
@@ -67,6 +63,6 @@ with gr.Blocks() as demo:
     gr.Markdown("## 🩺 AI Health Mentor — Dr. Alex")
     chatbot.render()
 
-# 🔹 Launch Space
+# 🔹 Launch
 if __name__ == "__main__":
     demo.launch(show_error=True)