import os import gradio as gr import requests import pandas as pd # Import your upgraded agent from agent import GeminiAgent # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" # This is the security gate. Only this user can run submissions. MY_HF_USERNAME = "benjipeng" def run_and_submit_all(profile: gr.OAuthProfile | None): """ Fetches all questions, runs the GeminiAgent on them, submits all answers, and displays the results. This function is restricted to a specific user and provides file context to the agent. """ # --- Determine HF Space Runtime URL and Repo URL --- space_id = os.getenv("SPACE_ID") # --- User Authentication and Authorization --- if not profile: return "Please Login to Hugging Face with the button to run the evaluation.", None username = profile.username print(f"User logged in: {username}") if username != MY_HF_USERNAME: print(f"Access denied for user: {username}. Allowed user is {MY_HF_USERNAME}.") return f"Error: This Space is configured for a specific user. Access denied for '{username}'.", None api_url = DEFAULT_API_URL questions_url = f"{api_url}/questions" submit_url = f"{api_url}/submit" # 1. Instantiate Agent print("Instantiating agent...") try: agent = GeminiAgent() except Exception as e: error_msg = f"Error initializing agent: {e}" print(error_msg) return error_msg, None agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" print(f"Code link for submission: {agent_code}") # 2. Fetch Questions print(f"Fetching questions from: {questions_url}") try: response = requests.get(questions_url, timeout=20) response.raise_for_status() questions_data = response.json() if not questions_data: print("Fetched questions list is empty.") return "Fetched questions list is empty or invalid format.", None print(f"Fetched {len(questions_data)} questions.") except requests.exceptions.RequestException as e: error_msg = f"Error fetching questions: {e}" print(error_msg) return error_msg, None except requests.exceptions.JSONDecodeError as e: error_msg = f"Error decoding server response for questions: {e}" print(error_msg) print(f"Response text: {response.text[:500]}") return error_msg, None # 3. Run your Agent (with context injection) results_log = [] answers_payload = [] print(f"Running agent on {len(questions_data)} questions...") for item in questions_data: task_id = item.get("task_id") question_text = item.get("question") # This is the key improvement: check if a file is associated with the question has_file = item.get("file", None) is not None if not task_id or question_text is None: print(f"Skipping item with missing task_id or question: {item}") continue # Modify the question to give the agent context about the file's existence if has_file: modified_question = f"{question_text}\n\n[Agent Note: A file is attached to this question. Use the 'read_file_from_api' tool to access it if needed.]" else: modified_question = question_text try: # Pass BOTH the modified question and the task_id to the agent submitted_answer = agent(modified_question, task_id) answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer}) results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer}) except Exception as e: print(f"Error running agent on task {task_id}: {e}") results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"}) if not answers_payload: return "Agent did not produce any answers to submit.", pd.DataFrame(results_log) # 4. Prepare Submission submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload} status_update = f"Agent finished. Submitting {len(answers_payload)} answers for user '{username}'..." print(status_update) # 5. Submit print(f"Submitting {len(answers_payload)} answers to: {submit_url}") try: response = requests.post(submit_url, json=submission_data, timeout=120) # Increased timeout response.raise_for_status() result_data = response.json() final_status = ( f"Submission Successful!\n" f"User: {result_data.get('username')}\n" f"Overall Score: {result_data.get('score', 'N/A')}% " f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n" f"Message: {result_data.get('message', 'No message received.')}" ) print("Submission successful.") results_df = pd.DataFrame(results_log) return final_status, results_df except requests.exceptions.HTTPError as e: error_detail = f"Server responded with status {e.response.status_code}." try: error_json = e.response.json() error_detail += f" Detail: {error_json.get('detail', e.response.text)}" except requests.exceptions.JSONDecodeError: error_detail += f" Response: {e.response.text[:500]}" status_message = f"Submission Failed: {error_detail}" print(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df except requests.exceptions.RequestException as e: status_message = f"Submission Failed: Network error - {e}" print(status_message) results_df = pd.DataFrame(results_log) return status_message, results_df # --- Build Gradio Interface using Blocks --- with gr.Blocks() as demo: gr.Markdown("# Gemini ReAct Agent for GAIA") gr.Markdown( """ **Instructions:** 1. Log in using the Hugging Face login button below. 2. Click 'Run Evaluation & Submit' to start the process. 3. The agent will fetch all 20 questions, reason about them step-by-step, use tools (like web search and a file reader), and submit the final answers for scoring. **Note:** This process can take several minutes. Please be patient. """ ) gr.LoginButton() run_button = gr.Button("Run Evaluation & Submit All Answers") status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False) results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True) run_button.click( fn=run_and_submit_all, outputs=[status_output, results_table] ) if __name__ == "__main__": print("\n" + "-"*30 + " App Starting " + "-"*30) demo.launch(debug=True, share=False)