Update app.py

app.py

@@ -1,34 +1,329 @@
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
+import tempfile
+from smolagents import (
+    CodeAgent,
+    DuckDuckGoSearchTool,
+    LiteLLMModel,
+    HfApiModel,
+    tool,
+)
 
-# (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Basic Agent Definition ---
-# ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-class BasicAgent:
+# --- GAIA System Prompt (adapted from official GAIA benchmark) ---
+GAIA_SYSTEM_PROMPT = """You are a general AI assistant. I will ask you a question. Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER].
+YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings.
+If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise.
+If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise.
+If you are asked for a comma separated list, apply the above rules depending on whether the element to be put in the list is a number or a string.
+
+IMPORTANT: Your FINAL ANSWER must be precise and concise. No extra words, no explanations after it.
+"""
+
+# --- Custom Tools ---
+
+@tool
+def visit_webpage(url: str) -> str:
+    """Visits a webpage at the given URL and returns its text content.
+    Use this to read articles, documentation, or any web page.
+
+    Args:
+        url: The full URL of the webpage to visit (must start with http:// or https://).
+    """
+    try:
+        import re
+        headers = {
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+        }
+        response = requests.get(url, headers=headers, timeout=30)
+        response.raise_for_status()
+
+        # Try to extract text from HTML
+        try:
+            from markdownify import markdownify
+            text = markdownify(response.text, heading_style="ATX")
+        except ImportError:
+            # Fallback: basic HTML tag removal
+            text = re.sub(r'<script[^>]*>.*?</script>', '', response.text, flags=re.DOTALL)
+            text = re.sub(r'<style[^>]*>.*?</style>', '', response.text, flags=re.DOTALL)
+            text = re.sub(r'<[^>]+>', ' ', text)
+            text = re.sub(r'\s+', ' ', text).strip()
+
+        # Truncate to avoid token limits
+        if len(text) > 15000:
+            text = text[:15000] + "\n\n[Content truncated - page too long]"
+        return text
+    except Exception as e:
+        return f"Error visiting {url}: {str(e)}"
+
+
+@tool
+def download_gaia_file(task_id: str) -> str:
+    """Downloads a file associated with a GAIA task and returns the local file path.
+    Use this when a question references an attached file that you need to read or analyze.
+
+    Args:
+        task_id: The task_id string for the GAIA question that has an associated file.
+    """
+    try:
+        api_url = DEFAULT_API_URL
+        file_url = f"{api_url}/files/{task_id}"
+        response = requests.get(file_url, timeout=30)
+        response.raise_for_status()
+
+        # Determine file extension from content-disposition or content-type
+        content_disp = response.headers.get("content-disposition", "")
+        content_type = response.headers.get("content-type", "")
+
+        ext = ".bin"
+        if "filename=" in content_disp:
+            import re
+            match = re.search(r'filename="?([^";\s]+)"?', content_disp)
+            if match:
+                fname = match.group(1)
+                if "." in fname:
+                    ext = "." + fname.rsplit(".", 1)[-1]
+        elif "text/plain" in content_type:
+            ext = ".txt"
+        elif "text/csv" in content_type:
+            ext = ".csv"
+        elif "application/json" in content_type:
+            ext = ".json"
+        elif "audio" in content_type:
+            ext = ".mp3"
+        elif "image/png" in content_type:
+            ext = ".png"
+        elif "image/jpeg" in content_type:
+            ext = ".jpg"
+        elif "application/pdf" in content_type:
+            ext = ".pdf"
+        elif "spreadsheet" in content_type or "excel" in content_type:
+            ext = ".xlsx"
+
+        # Save to temp file
+        tmp = tempfile.NamedTemporaryFile(delete=False, suffix=ext)
+        tmp.write(response.content)
+        tmp.close()
+
+        return f"File downloaded to: {tmp.name} (type: {content_type}, size: {len(response.content)} bytes)"
+    except Exception as e:
+        return f"Error downloading file for task {task_id}: {str(e)}"
+
+
+@tool
+def read_file_content(file_path: str) -> str:
+    """Reads and returns the content of a local file.
+    Supports text files (.txt, .csv, .json, .py, .md), and attempts to read Excel/PDF files.
+
+    Args:
+        file_path: The absolute path to the file to read.
+    """
+    try:
+        ext = file_path.rsplit(".", 1)[-1].lower() if "." in file_path else ""
+
+        if ext in ("txt", "csv", "json", "py", "md", "html", "xml", "log", "tsv"):
+            with open(file_path, "r", encoding="utf-8", errors="replace") as f:
+                content = f.read()
+            if len(content) > 20000:
+                content = content[:20000] + "\n\n[Content truncated]"
+            return content
+
+        elif ext in ("xlsx", "xls"):
+            import openpyxl
+            wb = openpyxl.load_workbook(file_path, data_only=True)
+            result = []
+            for sheet_name in wb.sheetnames:
+                ws = wb[sheet_name]
+                result.append(f"=== Sheet: {sheet_name} ===")
+                for row in ws.iter_rows(values_only=True):
+                    result.append("\t".join(str(c) if c is not None else "" for c in row))
+            return "\n".join(result)
+
+        elif ext == "pdf":
+            try:
+                import PyPDF2
+                with open(file_path, "rb") as f:
+                    reader = PyPDF2.PdfReader(f)
+                    text = []
+                    for page in reader.pages:
+                        text.append(page.extract_text() or "")
+                return "\n".join(text)
+            except ImportError:
+                return "PyPDF2 not available. Cannot read PDF files."
+
+        else:
+            # Try reading as binary and decode
+            with open(file_path, "rb") as f:
+                raw = f.read(5000)
+            try:
+                return raw.decode("utf-8")
+            except UnicodeDecodeError:
+                return f"Binary file ({ext}), size: {os.path.getsize(file_path)} bytes. Cannot display as text."
+
+    except Exception as e:
+        return f"Error reading file {file_path}: {str(e)}"
+
+
+@tool
+def wikipedia_search(query: str) -> str:
+    """Searches Wikipedia and returns a summary of the most relevant article.
+    Useful for factual questions about people, places, events, science, etc.
+
+    Args:
+        query: The search query to look up on Wikipedia.
+    """
+    try:
+        import wikipedia
+        results = wikipedia.search(query, results=3)
+        if not results:
+            return f"No Wikipedia results found for: {query}"
+
+        # Try to get the first result's summary
+        for result_title in results:
+            try:
+                page = wikipedia.page(result_title, auto_suggest=False)
+                summary = page.summary
+                if len(summary) > 5000:
+                    summary = summary[:5000] + "..."
+                return f"Wikipedia: {page.title}\n\n{summary}\n\nURL: {page.url}"
+            except (wikipedia.DisambiguationError, wikipedia.PageError):
+                continue
+
+        return f"Could not retrieve Wikipedia page for: {query}"
+    except Exception as e:
+        return f"Error searching Wikipedia: {str(e)}"
+
+
+@tool
+def perform_calculation(python_expression: str) -> str:
+    """Evaluates a mathematical Python expression and returns the result.
+    Use this for arithmetic, mathematical calculations, date computations, etc.
+
+    Args:
+        python_expression: A valid Python expression to evaluate (e.g. '2**10', 'round(3.14159, 2)', 'sum(range(1,101))').
+    """
+    try:
+        import math
+        import datetime
+        allowed_globals = {
+            "__builtins__": {},
+            "math": math,
+            "abs": abs, "round": round, "min": min, "max": max,
+            "sum": sum, "len": len, "sorted": sorted,
+            "int": int, "float": float, "str": str,
+            "list": list, "range": range, "enumerate": enumerate,
+            "zip": zip, "map": map, "filter": filter,
+            "pow": pow, "divmod": divmod,
+            "datetime": datetime,
+        }
+        result = eval(python_expression, allowed_globals)
+        return str(result)
+    except Exception as e:
+        return f"Calculation error: {str(e)}"
+
+
+# --- Agent Class ---
+class GaiaAgent:
     def __init__(self):
-        print("BasicAgent initialized.")
-    def __call__(self, question: str) -> str:
-        print(f"Agent received question (first 50 chars): {question[:50]}...")
-        fixed_answer = "This is a default answer."
-        print(f"Agent returning fixed answer: {fixed_answer}")
-        return fixed_answer
-
-def run_and_submit_all( profile: gr.OAuthProfile | None):
+        print("Initializing GaiaAgent...")
+
+        # Choose model - try LiteLLM with Anthropic first, fall back to HfApiModel
+        anthropic_key = os.getenv("ANTHROPIC_API_KEY")
+        openai_key = os.getenv("OPENAI_API_KEY")
+        hf_token = os.getenv("HF_TOKEN")
+
+        if anthropic_key:
+            print("Using Anthropic Claude via LiteLLM")
+            self.model = LiteLLMModel(
+                model_id="anthropic/claude-sonnet-4-20250514",
+                api_key=anthropic_key,
+                max_tokens=4096,
+                temperature=0.1,
+            )
+        elif openai_key:
+            print("Using OpenAI via LiteLLM")
+            self.model = LiteLLMModel(
+                model_id="openai/gpt-4o",
+                api_key=openai_key,
+                max_tokens=4096,
+                temperature=0.1,
+            )
+        else:
+            print("Using HfApiModel (free inference)")
+            self.model = HfApiModel(
+                model_id="Qwen/Qwen2.5-Coder-32B-Instruct",
+                token=hf_token,
+            )
+
+        # Build tools list
+        self.tools = [
+            DuckDuckGoSearchTool(),
+            visit_webpage,
+            download_gaia_file,
+            read_file_content,
+            wikipedia_search,
+            perform_calculation,
+        ]
+
+        # Create agent
+        self.agent = CodeAgent(
+            model=self.model,
+            tools=self.tools,
+            max_steps=12,
+            verbosity_level=1,
+            additional_authorized_imports=[
+                "json", "re", "math", "datetime", "collections",
+                "itertools", "functools", "statistics", "string",
+                "csv", "io", "os", "unicodedata",
+            ],
+        )
+
+        print("GaiaAgent initialized successfully.")
+
+    def __call__(self, question: str, task_id: str = None) -> str:
+        """Run the agent on a question and extract the final answer."""
+        print(f"\n{'='*60}")
+        print(f"Question (task_id={task_id}): {question[:100]}...")
+        print(f"{'='*60}")
+
+        # Build the prompt with task context
+        prompt = GAIA_SYSTEM_PROMPT + "\n\n"
+        if task_id:
+            prompt += f"[Note: This question has task_id='{task_id}'. If the question references an attached file, use the download_gaia_file tool with this task_id to get it.]\n\n"
+        prompt += f"Question: {question}"
+
+        try:
+            raw_answer = self.agent.run(prompt)
+            answer = str(raw_answer).strip()
+
+            # Extract just the final answer if the agent included "FINAL ANSWER:"
+            if "FINAL ANSWER:" in answer:
+                answer = answer.split("FINAL ANSWER:")[-1].strip()
+
+            # Clean up common artifacts
+            answer = answer.strip('"\'').strip()
+
+            print(f"Agent answer: {answer}")
+            return answer
+
+        except Exception as e:
+            print(f"Agent error: {e}")
+            return f"Error: {str(e)}"
+
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
-    Fetches all questions, runs the BasicAgent on them, submits all answers,
+    Fetches all questions, runs the GaiaAgent 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
+    space_id = os.getenv("SPACE_ID")
 
     if profile:
-        username= f"{profile.username}"
+        username = f"{profile.username}"
         print(f"User logged in: {username}")
     else:
         print("User not logged in.")
@@ -38,15 +333,15 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
 
-    # 1. Instantiate Agent ( modify this part to create your agent)
+    # 1. Instantiate Agent
     try:
-        agent = BasicAgent()
+        agent = GaiaAgent()
     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)
+    print(f"Agent code URL: {agent_code}")
 
     # 2. Fetch Questions
     print(f"Fetching questions from: {questions_url}")
@@ -55,51 +350,59 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         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
+            return "Fetched questions list is empty or invalid format.", None
         print(f"Fetched {len(questions_data)} questions.")
-    except requests.exceptions.RequestException as e:
+    except Exception 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
+    # 3. Run Agent on each question
     results_log = []
     answers_payload = []
     print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data:
+
+    for i, item in enumerate(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
+
+        print(f"\n--- Question {i+1}/{len(questions_data)} (task_id: {task_id}) ---")
         try:
-            submitted_answer = agent(question_text)
-            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})
+            submitted_answer = agent(question_text, task_id=task_id)
+            answers_payload.append({
+                "task_id": task_id,
+                "submitted_answer": submitted_answer,
+            })
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text[:100] + "..." if len(question_text) > 100 else 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}"})
+            print(f"Error running agent on task {task_id}: {e}")
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text[:100] + "..." if len(question_text) > 100 else 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}
+    # 4. Submit
+    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 = requests.post(submit_url, json=submission_data, timeout=120)
         response.raise_for_status()
         result_data = response.json()
         final_status = (
@@ -117,22 +420,12 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         try:
             error_json = e.response.json()
             error_detail += f" Detail: {error_json.get('detail', e.response.text)}"
-        except requests.exceptions.JSONDecodeError:
+        except Exception:
             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)
@@ -140,57 +433,47 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         return status_message, results_df
 
 
-# --- Build Gradio Interface using Blocks ---
+# --- Gradio Interface ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Basic Agent Evaluation Runner")
+    gr.Markdown("# 🤖 GAIA Benchmark Agent")
     gr.Markdown(
         """
-        **Instructions:**
+        **Agent powered by smolagents + CodeAgent**
 
-        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.
+        This agent uses web search, Wikipedia, file handling, and code execution to answer GAIA benchmark questions.
 
-        ---
-        **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.
+        **Instructions:**
+        1. Log in to your Hugging Face account below.
+        2. Click 'Run Evaluation & Submit All Answers' to start.
+        3. The agent will process all 20 questions (this may take several minutes).
+
+        **Tools available:** DuckDuckGo Search, Web Page Reader, Wikipedia, File Download/Reader, Python Calculator
         """
     )
 
     gr.LoginButton()
 
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
+    run_button = gr.Button("🚀 Run Evaluation & Submit All Answers", variant="primary")
 
     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]
+        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("\n" + "-" * 30 + " App Starting " + "-" * 30)
+    space_host = os.getenv("SPACE_HOST")
+    space_id = os.getenv("SPACE_ID")
 
-    print("-"*(60 + len(" App Starting ")) + "\n")
+    if space_host:
+        print(f"✅ SPACE_HOST: {space_host}")
+    if space_id:
+        print(f"✅ SPACE_ID: {space_id}")
+        print(f"   Repo: https://huggingface.co/spaces/{space_id}/tree/main")
 
-    print("Launching Gradio Interface for Basic Agent Evaluation...")
+    print("-" * 60)
+    print("Launching Gradio Interface...")
     demo.launch(debug=True, share=False)