import os import base64 import mimetypes import gradio as gr import requests import pandas as pd import anthropic # --- Constants --- DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space" MODEL = "claude-sonnet-4-6" # Optional manual answer overrides, keyed by task_id. Leave empty to let the # agent answer everything itself. (Previously this dict was referenced but # never defined, which crashed every call with a NameError.) CORRECT_ANSWERS = {} # GAIA is graded by normalized exact match, so the final answer must be bare — # no prose, no units, no "The answer is ...". We ask the model to reason, then # emit a single "FINAL ANSWER:" line that we extract before submitting. SYSTEM_PROMPT = """You are a general AI assistant answering questions from the GAIA benchmark. Work through the question step by step. Use the web_search tool to find current or factual information, and web_fetch to read specific pages you have found. If a file is attached, analyze it carefully. When you are confident, finish your response with a single final line in exactly this form: FINAL ANSWER: [YOUR FINAL ANSWER] Rules for YOUR FINAL ANSWER (it is graded by exact string match, so follow them precisely): - It must be a number, OR as few words as possible, OR a comma-separated list of numbers/strings. - Do not include units ($, %, USD, km, ...) unless the question explicitly asks for them. - Write numbers as digits with no thousands separators (e.g. 1000000, not 1,000,000). - Do not use articles or abbreviations for strings unless the question asks for them. - For a comma-separated list, put exactly one space after each comma. - Output nothing after the FINAL ANSWER line. """ # Server-side tools — these execute on Anthropic's infrastructure; no client loop needed. TOOLS = [ {"type": "web_search_20260209", "name": "web_search"}, {"type": "web_fetch_20260209", "name": "web_fetch"}, ] IMAGE_EXTS = {".png", ".jpg", ".jpeg", ".gif", ".webp"} TEXT_EXTS = {".txt", ".md", ".py", ".js", ".csv", ".json", ".xml", ".html", ".tsv"} # --- Agent Definition --- class BasicAgent: def __init__(self): # Reads ANTHROPIC_API_KEY from the environment (set it as a Space secret). print(f"Initializing Claude agent ({MODEL})...") self.client = anthropic.Anthropic() def _fetch_file_block(self, task_id, file_name): """Download a question's attached file and turn it into a content block.""" if not task_id: return None url = f"{DEFAULT_API_URL}/files/{task_id}" try: resp = requests.get(url, timeout=30) resp.raise_for_status() except requests.exceptions.RequestException as e: print(f"Could not fetch file for task {task_id}: {e}") return None ext = os.path.splitext(file_name or "")[1].lower() media_type = mimetypes.guess_type(file_name or "")[0] if ext in IMAGE_EXTS: data = base64.standard_b64encode(resp.content).decode("utf-8") return { "type": "image", "source": {"type": "base64", "media_type": media_type or "image/png", "data": data}, } if ext == ".pdf": data = base64.standard_b64encode(resp.content).decode("utf-8") return { "type": "document", "source": {"type": "base64", "media_type": "application/pdf", "data": data}, } if ext in TEXT_EXTS: try: text = resp.content.decode("utf-8", errors="replace") except Exception: text = "" return {"type": "text", "text": f"Attached file '{file_name}':\n\n{text}"} # Unsupported binary type (e.g. .xlsx, .mp3) — note it so the model can adapt. print(f"Unsupported attached file type for task {task_id}: {file_name}") return {"type": "text", "text": f"(An unsupported file '{file_name}' is attached but could not be read directly.)"} def _run_agent(self, messages): """Drive the agent loop, handling server-tool pause_turn continuations.""" resp = None for _ in range(12): resp = self.client.messages.create( model=MODEL, max_tokens=4096, system=SYSTEM_PROMPT, tools=TOOLS, messages=messages, ) # Server-tool loop hit its internal iteration cap; re-send to resume. if resp.stop_reason == "pause_turn": messages.append({"role": "assistant", "content": resp.content}) continue break if resp is None: return "" return "".join(b.text for b in resp.content if b.type == "text") @staticmethod def _extract_final(text): marker = "FINAL ANSWER:" idx = text.rfind(marker) if idx != -1: return text[idx + len(marker):].strip() # Fallback: last non-empty line. lines = [ln.strip() for ln in text.splitlines() if ln.strip()] return lines[-1] if lines else text.strip() def __call__(self, question: str, task_id: str = None, file_name: str = None) -> str: if task_id and task_id in CORRECT_ANSWERS: answer = CORRECT_ANSWERS[task_id] print(f"Using predefined answer for task {task_id}: {answer}") return answer print(f"Agent received question (first 80 chars): {question[:80]}...") content = [] if file_name: block = self._fetch_file_block(task_id, file_name) if block: content.append(block) content.append({"type": "text", "text": question}) try: raw = self._run_agent([{"role": "user", "content": content}]) answer = self._extract_final(raw) except Exception as e: print(f"Error generating answer: {e}") answer = "Error generating answer." print(f"Agent returning answer: {answer}") return answer def run_and_submit_all( profile: gr.OAuthProfile | None): """ Fetches all questions, runs the BasicAgent on them, submits all answers, and displays the results. """ # --- Determine HF Space Runtime URL and Repo URL --- space_id = os.getenv("SPACE_ID") # Get the SPACE_ID for sending link to the code if profile: username= f"{profile.username}" print(f"User logged in: {username}") else: print("User not logged in.") return "Please Login to Hugging Face with the button.", None api_url = DEFAULT_API_URL questions_url = f"{api_url}/questions" submit_url = f"{api_url}/submit" # 1. Instantiate Agent ( modify this part to create your agent) try: agent = BasicAgent() except Exception as e: print(f"Error instantiating agent: {e}") return f"Error initializing agent: {e}", None # In the case of an app running as a hugging Face space, this link points toward your codebase ( usefull for others so please keep it public) agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" print(agent_code) # 2. Fetch Questions print(f"Fetching questions from: {questions_url}") try: response = requests.get(questions_url, timeout=15) response.raise_for_status() questions_data = response.json() questions_data = [q for q in questions_data if q.get("Level") == "1"] print(f"Filtered to {len(questions_data)} Level 1 questions.") 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: print(f"Error fetching questions: {e}") return f"Error fetching questions: {e}", None except requests.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 occurred fetching questions: {e}") return f"An unexpected error occurred fetching questions: {e}", None # 3. Run your Agent 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") file_name = item.get("file_name") Level = item.get("Level", "?") print(f"\n➡️ Task {task_id} (Level {Level})") print(f"Question: {question_text[:100]}...") 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, task_id=task_id, file_name=file_name) 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: print("Agent did not produce any answers to submit.") 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=60) 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.Timeout: status_message = "Submission Failed: The request timed out." 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 except Exception as e: status_message = f"An unexpected error occurred during submission: {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("# Basic Agent Evaluation Runner") gr.Markdown( """ **Instructions:** 1. Please clone this space, then modify the code to define your agent's logic, the tools, the necessary packages, etc ... 2. Log in to your Hugging Face account using the button below. This uses your HF username for submission. 3. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score. --- **Disclaimers:** Once clicking on the "submit button, it can take quite some time ( this is the time for the agent to go through all the questions). This space provides a basic setup and is intentionally sub-optimal to encourage you to develop your own, more robust solution. For instance for the delay process of the submit button, a solution could be to cache the answers and submit in a seperate action or even to answer the questions in async. """ ) gr.LoginButton() run_button = gr.Button("Run Evaluation & Submit All Answers") status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False) # Removed max_rows=10 from DataFrame constructor 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) # Check for SPACE_HOST and SPACE_ID at startup for information space_host_startup = os.getenv("SPACE_HOST") space_id_startup = os.getenv("SPACE_ID") # Get SPACE_ID at startup if space_host_startup: print(f"✅ SPACE_HOST found: {space_host_startup}") print(f" Runtime URL should be: https://{space_host_startup}.hf.space") else: print("ℹ️ SPACE_HOST environment variable not found (running locally?).") if space_id_startup: # Print repo URLs if SPACE_ID is found print(f"✅ SPACE_ID found: {space_id_startup}") print(f" Repo URL: https://huggingface.co/spaces/{space_id_startup}") print(f" Repo Tree URL: https://huggingface.co/spaces/{space_id_startup}/tree/main") else: print("ℹ️ SPACE_ID environment variable not found (running locally?). Repo URL cannot be determined.") print("-"*(60 + len(" App Starting ")) + "\n") print("Launching Gradio Interface for Basic Agent Evaluation...") demo.launch(debug=True, share=False)