Update app.py

app.py

@@ -1,53 +1,71 @@
+import os
 import gradio as gr
 from huggingface_hub import InferenceClient
 
 
 def respond(
-    message,
+    message: str,
     history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
     hf_token: gr.OAuthToken,
 ):
     """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
+    Generate a response using the HuggingFace Inference API.
+
+    The system prompt is taken from the secret **prec_chat** (defined in the
+    Gradio secrets file).  Users can no longer edit the system prompt.
     """
+    # ----------------------------------------------------------------------
+    # 1️⃣  Load the system prompt from the secret.
+    # ----------------------------------------------------------------------
+    # If the secret is missing we fall back to a generic prompt so the app
+    # still works locally.
+    system_message = os.getenv("prec_chat", "You are a helpful assistant.")
+
+    # ----------------------------------------------------------------------
+    # 2️⃣  Initialise the HF inference client.
+    # ----------------------------------------------------------------------
     client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
 
+    # ----------------------------------------------------------------------
+    # 3️⃣  Build the message list for the chat completion endpoint.
+    # ----------------------------------------------------------------------
     messages = [{"role": "system", "content": system_message}]
+    messages.extend(history)                     # previous conversation
+    messages.append({"role": "user", "content": message})  # current query
 
-    messages.extend(history)
-
-    messages.append({"role": "user", "content": message})
-
+    # ----------------------------------------------------------------------
+    # 4️⃣  Stream the response back to the UI.
+    # ----------------------------------------------------------------------
     response = ""
-
-    for message in client.chat_completion(
+    for chunk in client.chat_completion(
         messages,
         max_tokens=max_tokens,
         stream=True,
         temperature=temperature,
         top_p=top_p,
     ):
-        choices = message.choices
+        # The API returns a list of choices – we only care about the first one.
+        choices = chunk.choices
         token = ""
-        if len(choices) and choices[0].delta.content:
+        if choices and choices[0].delta.content:
             token = choices[0].delta.content
 
         response += token
         yield response
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# --------------------------------------------------------------------------
+# UI definition – the system‑prompt textbox has been removed.
+# --------------------------------------------------------------------------
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        # System prompt is now fixed via the secret, so we only expose the
+        # generation parameters.
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(
@@ -55,7 +73,7 @@ chatbot = gr.ChatInterface(
             maximum=1.0,
             value=0.95,
             step=0.05,
-            label="Top-p (nucleus sampling)",
+            label="Top‑p (nucleus sampling)",
         ),
     ],
 )