Update app.py

app.py

@@ -59,8 +59,9 @@ def chat_fn(message, history):
         {
             "role": "system",
             "content": (
-                "You are a helpful, precise assistant. Use the provided context to answer when available."
-                " If no relevant context is found, reply politely or ask the user to rephrase."
+                "You are a precise, concise assistant. Use ONLY the provided context."
+                " Respond in 1–2 short sentences maximum. Do not explain unless asked."
+                " If unsure, say: 'I couldn’t find the answer based on available context.'"
                 "\n\nContext:\n" + context
             )
         }
@@ -70,12 +71,13 @@ def chat_fn(message, history):
         messages.append({"role": "user", "content": user})
         messages.append({"role": "assistant", "content": assistant})
 
-    messages.append({"role": "user", "content": message})
+    # ✅ Encourage brevity in user input
+    messages.append({"role": "user", "content": message + "\n\nPlease answer briefly."})
 
     payload = {
         "model": MODEL,
         "messages": messages,
-        "max_tokens": 300,
+        "max_tokens": 150,
     }
 
     try:
@@ -109,7 +111,7 @@ async def chat_api(request: Request):
 demo = gr.ChatInterface(
     fn=chat_fn,
     title="💬 CODEX MIRXA KAMRAN",
-    description="Ask anything related to our knowledge base. Fast, short and precise responses!",
+    description="Ask anything related to our knowledge base. Short, fast and precise answers!",
     theme="soft"
 )