Update app.py

app.py

@@ -26,6 +26,12 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
 # --- Basic Agent Definition ---
 # ---- THIS IS WHERE YOU CAN BUILD WHAT YOU WANT ----
+# This class is a ready-to-run wrapper that:
+# 1. Authenticates to the Hub
+# 2. Spins up a server-side Qwen-32B LLM.
+# 3. Gives it a DuckDuckGo search plug-in plus smolagents' standard library
+# 4. Primes it with strict grading instructions.
+# 5. Exposes a clean, callable interface for what  ever frontend(Gradio, FastAPI, etc.) you bolt on.
 class BasicAgent:
     def __init__(self):
         # Pull a HF access token from the Space's secrets or your local shell. You can download private models, call paid-tier Inference endpoints, push artefacts
@@ -62,9 +68,12 @@ class BasicAgent:
         tools=[self.search_tool],
         # drops in a small standard library (Python REPL, JSON loader etc.) so you can solve many tasks without defining anything else.
         add_base_tools=True # - python_repl, browser, math etc.
+        # CodeAgent's auto_document_tools convenience flag 
+        auto_document_tools=True
     )
 
     # Send a single "bootstrap" run whose only job is lock in behaviour rules:
+    # The returned text is captured in self.responses.
     self.response = self.agent.run(
         """
         You are a general AI assistant.
@@ -76,4 +85,111 @@ class BasicAgent:
         You have access to the following tools:
         Tool Name: search_tool, description:  lets you search and browse the internet for accessing the most updated information out there.
         If you require more tools to get a correct answer, create your own tools to utilize.    
-        """)
+        """)
+    # Turning BasicAgent into a callable object 
+    # It means you can drop it straight into Gradio (or any other framework) without wrapping it in a standalone function.
+    # Debug prints show the round-trip in the server logs.
+    def __call__(self, question: str) -> str:
+        print(f"Agent received question:")
+        response = self.agent.run(question)
+        # the reply is generated on-the-fly, not hard coded.
+        print(f"Agent returning answer: {response}")
+        return response
+
+# 1. Check if the user is logged in
+# 2. Download questions from a grading API.
+# 3. Use the BasicAgent to generate answers
+# 4. Submit those answers back to the API.
+# 5. Return the grading results + a full log for UI display (e.g. Gradio Table)
+# Includes detailed logging, robust error handling, and submission payload formatting
+def run_and_submit_all( profile: gr.OAuthProfile | None):
+    """
+    Fetches all questions, runs the BasicAgent on them, submits all answers, and display the results.
+    """
+    # --- Determine HF Space Runtime URL and Repo URL ---
+    # Authenticate user and runtime info
+    # Grabbing space_id from the environment lets the app dynamically construct a URL to your codebase.
+    # This will be included in the submission for transparency (important in peer-review courses.)
+    space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code
+
+    # If the gradio OAuth profile object is present, extract the username.
+    if profile:
+        username = f"{profile.username}"
+        print(f"User logged in: {username}")
+    # Otherwise, early exit with a friendly error message
+    else:
+        print("User not logged in.")
+        return "Please login to Hugging Face with the button.",None
+
+    # --- PrePare API endpoints ---
+    # Uses the provided scoring end point (defaulting to the course's hosted  backen)
+    # Constucts two URLs:
+    api_url = DEFAULT_API_URL
+    # URL to Fetch the question bank.
+    question_url = f"{api_url}/questions"
+    # URL to POST answers for grading
+    submit_url = f"{api_url}/submit"
+
+    # 1. Instantiate Agent ( modify this part to create your agent)
+    # Tries to spin up your BasicAgent class from earlier.
+    # Includes token validation, model loading, tool setup, and system prompt injection.
+    # If this fails, the app gracefully exits, returning a user-visible error.
+    try:
+        agent = BasicAgent()
+    except Exception as e:
+        print(f"Error instantiating agent: {e}")
+        return f"Error initialiazing agent: {e}", None
+
+    # In the case of an app running as a HF space, this link points toward your codebase 
+    # (usefull for others so please keep it public)
+    # Builds a link to your code repor on HF Hub (public space)
+    agent_code =  f"https://huggingface.co/spaces/{space_id}/tree/main"
+    # Gets submitted with the answers for transparacey
+    print(agent_code)
+
+    # 2. Fetch Questions
+    # --- FETCH QUESTIONS FROM THE BACKEND ---
+    print(f"Fetching questions from: {questions_url}")
+    # Tries to GET the questions from the course's scoring server
+    try:
+        # Timout and error handling ensure the app does not hang or crash.
+        
+        response = requests.get(requests, timeout=15)
+        questions_data = response.json()
+        # handles edge cases like empty response, malformed JSON, network Errors
+        # Empty response handling
+        if not questions_data:
+            print("Fetched questions list is empty.")
+            return "Fetched questions list is empty or invalid format.", None
+        print(f"Fetched {len(question_data) questions.}")
+    except requests.exceptions,RequestException as e:
+        print(f"Error fetching questions: {e}")
+        return f"Error fetching questions: {e}", None
+    except reqests.exceptions.JSONDecodeError as e:
+        print(f"Error decoding JSON response from questions endpoint: {e}")
+        print(f"Response text: {response.text[:500]}")
+        return f"Error decoding server response for questions: {e}, None"
+    except Exception as e: 
+        print(f"An unexpected error occured fetching questions: {e}")
+        return f"An unexpected error occurred fetching questions: {e}", None
+    # 3. Run your agent.
+    # Loop through questions and generate answers
+    results_log = [] # Used to make a DataFrame for UI display (question + answer)
+    answers_payload = [] # sent to grading API in the final submission
+
+    for item in questions_data:
+        task_id = item.get("task_id")
+        question_text = item.get("question")
+        if not task_id or question_text is None:
+            print(f"Skipping item with missing task_id or question: {item}")
+            continue
+        try: 
+            submitted_answer = agent(question_text)
+            answers_payload.append({"task_id": task_id, "submmitted_answer": submitted_answer})
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
+        except Exception as e:
+            print(f"Erron running agent on task {task_id}: {e}")
+            results_log.append
+    
+
+