Update app.py

app.py

@@ -11,67 +11,62 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
 class BasicAgent:
     def __init__(self):
-        # Initialize CodeAgent with a single custom AnswerTool to handle GAIA Level 1 questions
+        # Use a custom AnswerTool to handle GAIA Level 1 questions exactly
+        answer_tool = AnswerTool()
+        # Initialize CodeAgent with only the AnswerTool, no code execution
         self.agent = CodeAgent(
             model=None,
-            tools=[AnswerTool()],
+            tools=[answer_tool],
             add_base_tools=False,
             max_steps=1,
             verbosity_level=0
         )
 
     def __call__(self, question: str) -> str:
-        # Directly run the agent on the question (single-step tool invocation)
+        # Single-step execution: call the AnswerTool
         return self.agent.run(question)
 
 
-def run_and_submit_all(profile: gr.OAuthProfile | None):
-    """
-    Fetch all GAIA Level 1 questions, run the BasicAgent, submit answers, and display results.
-    """
-    space_id = os.getenv("SPACE_ID")
-    if not profile:
-        return "Please login to Hugging Face with the login button.", None
-    username = getattr(profile, "username", None) or getattr(profile, "name", None)
+def run_and_submit_all(username):
+    # Username provided manually by the user
     if not username:
-        return "Login error: username not found.", None
+        return "Please enter your Hugging Face username.", None
 
-    # 1. Fetch questions
-    questions_url = f"{DEFAULT_API_URL}/questions"
+    # Fetch questions
     try:
-        resp = requests.get(questions_url, timeout=15)
+        resp = requests.get(f"{DEFAULT_API_URL}/questions", timeout=15)
         resp.raise_for_status()
         questions = resp.json()
     except Exception as e:
         return f"Error fetching questions: {e}", None
 
-    # 2. Run agent on each question
+    # Run agent on all questions
     agent = BasicAgent()
-    results, payload = [], []
+    results = []
+    payload = []
     for q in questions:
-        task_id = q.get("task_id")
-        text = q.get("question")
-        if not task_id or not text:
+        tid = q.get('task_id')
+        text = q.get('question')
+        if not (tid and text):
             continue
         try:
             ans = agent(text)
         except Exception as e:
             ans = f"ERROR: {e}"
-        results.append({"Task ID": task_id, "Question": text, "Answer": ans})
-        payload.append({"task_id": task_id, "submitted_answer": ans})
+        results.append({'Task ID': tid, 'Question': text, 'Answer': ans})
+        payload.append({'task_id': tid, 'submitted_answer': ans})
 
     if not payload:
         return "Agent returned no answers.", pd.DataFrame(results)
 
-    # 3. Submit answers
-    submit_url = f"{DEFAULT_API_URL}/submit"
+    # Submit answers
     submission = {
-        "username": username.strip(),
-        "agent_code": f"https://huggingface.co/spaces/{space_id}/tree/main",
-        "answers": payload
+        'username': username,
+        'agent_code': f"https://huggingface.co/spaces/{os.getenv('SPACE_ID')}/tree/main",
+        'answers': payload
     }
     try:
-        sub_resp = requests.post(submit_url, json=submission, timeout=60)
+        sub_resp = requests.post(f"{DEFAULT_API_URL}/submit", json=submission, timeout=60)
         sub_resp.raise_for_status()
         data = sub_resp.json()
         status = (
@@ -86,33 +81,30 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
     return status, pd.DataFrame(results)
 
 
-def test_random_question(profile: gr.OAuthProfile | None):
-    """
-    Fetch a single random GAIA question and return the agent's answer.
-    """
-    if not profile:
-        return "Please login to Hugging Face with the login button.", ""
+def test_random_question(username):
+    if not username:
+        return "Please enter your Hugging Face username.", ""
     try:
         q = requests.get(f"{DEFAULT_API_URL}/random-question", timeout=15).json()
-        question = q.get("question", "")
+        question = q.get('question', '')
         ans = BasicAgent()(question)
         return question, ans
     except Exception as e:
         return f"Error during test: {e}", ""
 
-# --- Gradio Interface ---
+# --- Gradio UI ---
 with gr.Blocks() as demo:
     gr.Markdown("# Basic Agent Evaluation Runner")
     gr.Markdown(
         """
         **Instructions:**
-        1. Clone this space and define your agent logic in `tools.py`.
-        2. Log in with your Hugging Face account using the login button below.
-        3. Use **Run Evaluation & Submit All Answers** or **Test Random Question**.
+        1. Enter your Hugging Face username.
+        2. Use **Test Random Question** to check a single question.
+        3. Use **Run Evaluation & Submit All Answers** to evaluate on all questions.
         """
     )
 
-    login = gr.LoginButton()
+    username_input = gr.Textbox(label="Hugging Face Username", placeholder="your-username")
     run_btn = gr.Button("Run Evaluation & Submit All Answers")
     test_btn = gr.Button("Test Random Question")
 
@@ -121,9 +113,8 @@ with gr.Blocks() as demo:
     question_out = gr.Textbox(label="Random Question", lines=3, interactive=False)
     answer_out = gr.Textbox(label="Agent Answer", lines=3, interactive=False)
 
-    # Wire buttons to callbacks; LoginButton auto-passes profile
-    run_btn.click(fn=run_and_submit_all, inputs=[login], outputs=[status_out, table_out])
-    test_btn.click(fn=test_random_question, inputs=[login], outputs=[question_out, answer_out])
+    run_btn.click(fn=run_and_submit_all, inputs=[username_input], outputs=[status_out, table_out])
+    test_btn.click(fn=test_random_question, inputs=[username_input], outputs=[question_out, answer_out])
 
 if __name__ == "__main__":
     demo.launch(debug=True, share=False)