initial commit of agent with score of 60

.env.example

@@ -0,0 +1 @@
+XAI_API_KEY=your-groq-api-key-here

.gitignore

@@ -0,0 +1,12 @@
+.env
+env.example
+gaia-env/*
+/gaia-env
+project_planning.md
+projectdescription.md
+test_agent.py
+test_groq_api.py
+test_groq_api_with_dotenv.py
+test_results.json
+test_xai_api.py
+update_groq_key.py

README.md

@@ -1,13 +1,163 @@
 ---
-title: AgentCourseFinalProject
-emoji: 🌍
-colorFrom: purple
-colorTo: green
+title: GAIA Agent for Hugging Face Agents Course
+emoji: 🕵🏻‍♂️
+colorFrom: indigo
+colorTo: indigo
 sdk: gradio
-sdk_version: 5.26.0
+sdk_version: 5.25.2
 app_file: app.py
 pinned: false
-short_description: Agent Code for Final Project
+hf_oauth: true
+# optional, default duration is 8 hours/480 minutes. Max duration is 30 days/43200 minutes.
+hf_oauth_expiration_minutes: 480
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# GAIA Agent for Hugging Face Agents Course
+
+This project implements a powerful intelligent agent using the SmolAgents framework to tackle the GAIA benchmark questions for the Hugging Face Agents course final assessment.
+
+## Project Overview
+
+The GAIA benchmark consists of challenging questions that require an agent to use various tools, including web search, file processing, and reasoning capabilities. This agent is designed to:
+
+1. Receive questions from the GAIA API
+2. Process and understand the questions
+3. Use appropriate tools to find answers
+4. Format and return precise answers
+
+## Features
+
+- **SmolAgents Integration**: Uses CodeAgent for flexible problem-solving with Python code execution
+- **Multi-Model Support**: 
+  - Compatible with Hugging Face models
+  - OpenAI models (GPT-4o and others)
+  - X.AI's Grok models
+  - Anthropic, Cohere, and Mistral models via LiteLLM
+- **Enhanced Tool Suite**: 
+  - Web search via DuckDuckGo
+  - Python interpreter for code execution
+  - File handling (reading, saving, downloading)
+  - Data analysis for CSV and Excel files
+  - Image processing with OCR capabilities (when available)
+- **Flexible Environment Configuration**: 
+  - Easy setup via environment variables or .env file
+  - Fallback mechanisms for missing dependencies
+  - Support for both local and secure E2B code execution
+- **Answer Processing**: 
+  - Special handling for reversed text questions
+  - Precise answer formatting for benchmark submission
+  - Automatic cleanup of model responses for exact matching
+- **Interactive UI**: Gradio interface for running the agent and submitting answers
+
+## Setup
+
+### Prerequisites
+
+- Python 3.8+
+- Hugging Face account
+- API keys for your preferred models (HuggingFace, OpenAI, X.AI, etc.)
+
+### Installation
+
+1. Clone this repository
+2. Install the required dependencies:
+
+```bash
+pip install -r requirements.txt
+```
+
+3. Copy the example environment file and add your API keys:
+
+```bash
+cp env.example .env
+# Edit .env with your API keys and configuration
+```
+
+### Configuration
+
+Configure the agent by setting these environment variables or editing the `.env` file:
+
+#### API Keys
+```
+HUGGINGFACEHUB_API_TOKEN=your_huggingface_token_here
+OPENAI_API_KEY=your_openai_key_here
+XAI_API_KEY=your_xai_api_key_here  # For X.AI/Grok models
+```
+
+#### Agent Configuration
+```
+AGENT_MODEL_TYPE=OpenAIServerModel  # HfApiModel, InferenceClientModel, LiteLLMModel, OpenAIServerModel
+AGENT_MODEL_ID=gpt-4o  # Model ID depends on the model type
+AGENT_TEMPERATURE=0.2
+AGENT_EXECUTOR_TYPE=local  # local or e2b for secure execution
+AGENT_VERBOSE=true  # Set to true for detailed logging
+```
+
+#### Advanced Configuration
+```
+AGENT_PROVIDER=hf-inference  # Provider for InferenceClientModel
+AGENT_TIMEOUT=120  # Timeout in seconds for API calls
+AGENT_API_BASE=https://api.groq.com/openai/v1  # For X.AI when using OpenAIServerModel
+```
+
+## Usage
+
+### Running the Agent
+
+Launch the Gradio interface with:
+
+```bash
+python app.py
+```
+
+Then:
+1. Log in to your Hugging Face account using the button in the interface
+2. Click "Run Evaluation & Submit All Answers"
+
+### Testing
+
+To test the agent with sample questions before running the full evaluation:
+
+```bash
+python test_agent.py
+```
+
+For more focused testing with specific APIs:
+
+```bash
+python test_groq_api.py  # Test X.AI/Groq API integration
+python test_xai_api.py  # Test X.AI API integration
+```
+
+## Project Structure
+
+- `app.py`: Main application with Gradio interface
+- `core_agent.py`: Agent implementation with SmolAgents framework
+- `api_integration.py`: Client for interacting with GAIA API
+- `test_agent.py`: Testing script with sample questions
+- `test_groq_api.py` & `test_xai_api.py`: API-specific test scripts
+- `update_groq_key.py`: Utility for updating API keys
+- `project_planning.md`: Development roadmap and progress tracking
+- `requirements.txt`: Project dependencies
+
+## Tools Implementation
+
+The agent includes several custom tools:
+
+1. **save_and_read_file**: Save content to a temporary file and return the path
+2. **download_file_from_url**: Download a file from a URL and save it locally
+3. **extract_text_from_image**: OCR for extracting text from images (requires pytesseract)
+4. **analyze_csv_file**: Load and analyze CSV files using pandas
+5. **analyze_excel_file**: Load and analyze Excel files using pandas
+
+## Resources
+
+- [GAIA Benchmark Information](https://huggingface.co/spaces/gaia-benchmark/leaderboard)
+- [SmolAgents Documentation](https://huggingface.co/docs/smolagents/en/index)
+- [Hugging Face Agents Course](https://huggingface.co/agents-course)
+
+## License
+
+This project is licensed under the MIT License.

api_integration.py

@@ -0,0 +1,39 @@
+import requests
+from typing import List, Dict, Any
+from core_agent import GAIAAgent
+
+class GAIAApiClient:
+    def __init__(self, api_url="https://agents-course-unit4-scoring.hf.space"):
+        self.api_url = api_url
+        self.questions_url = f"{api_url}/questions"
+        self.submit_url = f"{api_url}/submit"
+        self.files_url = f"{api_url}/files"
+        
+    def get_questions(self) -> List[Dict[str, Any]]:
+        """Fetch all evaluation questions"""
+        response = requests.get(self.questions_url)
+        response.raise_for_status()
+        return response.json()
+    
+    def get_random_question(self) -> Dict[str, Any]:
+        """Fetch a single random question"""
+        response = requests.get(f"{self.api_url}/random-question")
+        response.raise_for_status()
+        return response.json()
+    
+    def get_file(self, task_id: str) -> bytes:
+        """Download a file for a specific task"""
+        response = requests.get(f"{self.files_url}/{task_id}")
+        response.raise_for_status()
+        return response.content
+    
+    def submit_answers(self, username: str, agent_code: str, answers: List[Dict[str, Any]]) -> Dict[str, Any]:
+        """Submit agent answers and get score"""
+        data = {
+            "username": username,
+            "agent_code": agent_code,
+            "answers": answers
+        }
+        response = requests.post(self.submit_url, json=data)
+        response.raise_for_status()
+        return response.json()

app.py

@@ -0,0 +1,279 @@
+import os
+import gradio as gr
+import requests
+import inspect
+import pandas as pd
+from core_agent import GAIAAgent
+
+# (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:
+    def __init__(self):
+        print("BasicAgent initialized.")
+        
+        # Initialize the GAIAAgent with local execution
+        try:
+            # Load environment variables if dotenv is available
+            try:
+                import dotenv
+                dotenv.load_dotenv()
+                print("Loaded environment variables from .env file")
+            except ImportError:
+                print("python-dotenv not installed, continuing with environment as is")
+            
+            # Try to load API keys from environment
+            api_key = os.getenv("XAI_API_KEY") or os.getenv("OPENAI_API_KEY") or os.getenv("HUGGINGFACEHUB_API_TOKEN")
+            
+            # If we have at least one API key, use a model-based approach
+            if api_key:
+                # Default model parameters
+                model_type = os.getenv("AGENT_MODEL_TYPE", "OpenAIServerModel")
+                model_id = os.getenv("AGENT_MODEL_ID", "gpt-4o")
+                
+                if os.getenv("XAI_API_KEY"):
+                    # Use X.AI API with OpenAIServerModel
+                    self.gaia_agent = GAIAAgent(
+                        model_type="OpenAIServerModel",
+                        model_id="grok-3-latest",  # X.AI's model
+                        api_key=os.getenv("XAI_API_KEY"),
+                        api_base="https://api.x.ai/v1",  # X.AI's endpoint, not Groq
+                        temperature=0.2,
+                        executor_type="local",
+                        verbose=False
+                    )
+                    print("Using OpenAIServerModel with X.AI API")
+                elif model_type == "HfApiModel" and os.getenv("HUGGINGFACEHUB_API_TOKEN"):
+                    # Use Hugging Face API
+                    self.gaia_agent = GAIAAgent(
+                        model_type="HfApiModel",
+                        model_id=model_id,
+                        api_key=os.getenv("HUGGINGFACEHUB_API_TOKEN"),
+                        temperature=0.2,
+                        executor_type="local",
+                        verbose=False
+                    )
+                    print(f"Using HfApiModel with model_id: {model_id}")
+                else:
+                    # Default to OpenAI API
+                    self.gaia_agent = GAIAAgent(
+                        model_type="OpenAIServerModel",
+                        model_id=model_id,
+                        api_key=os.getenv("OPENAI_API_KEY"),
+                        temperature=0.2,
+                        executor_type="local",
+                        verbose=False
+                    )
+                    print(f"Using OpenAIServerModel with model_id: {model_id}")
+            else:
+                # No API keys available, use a local model setup with minimal dependencies
+                self.gaia_agent = GAIAAgent(
+                    model_type="HfApiModel",
+                    model_id="gpt2",  # Simple model for basic testing
+                    temperature=0.2,
+                    executor_type="local",
+                    verbose=False
+                )
+                print("Warning: No API keys found. Using a basic local execution setup.")
+                
+        except Exception as e:
+            print(f"Error initializing GAIAAgent: {e}")
+            self.gaia_agent = None
+            print("WARNING: Failed to initialize agent. Falling back to basic responses.")
+            
+    def __call__(self, question: str) -> str:
+        print(f"Agent received question (first 50 chars): {question[:50]}...")
+        
+        # Check if we have a functioning GAIA agent
+        if self.gaia_agent:
+            try:
+                # Process the question using the GAIA agent
+                answer = self.gaia_agent.answer_question(question)
+                print(f"Agent generated answer: {answer[:50]}..." if len(answer) > 50 else f"Agent generated answer: {answer}")
+                return answer
+            except Exception as e:
+                print(f"Error processing question: {e}")
+                # Fall back to a simple response on error
+                return "An error occurred while processing your question. Please check the agent logs for details."
+        else:
+            # We don't have a valid agent, provide a basic response
+            return "The agent is not properly initialized. Please check your API keys and configuration."
+
+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()
+        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")
+        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, "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)

core_agent.py

@@ -0,0 +1,492 @@
+from smolagents import (
+    CodeAgent, 
+    DuckDuckGoSearchTool, 
+    HfApiModel, 
+    LiteLLMModel,
+    OpenAIServerModel,
+    PythonInterpreterTool,
+    tool,
+    InferenceClientModel
+)
+from typing import List, Dict, Any, Optional
+import os
+import tempfile
+import re
+import json
+import requests
+from urllib.parse import urlparse
+
+@tool
+def save_and_read_file(content: str, filename: Optional[str] = None) -> str:
+    """
+    Save content to a temporary file and return the path.
+    Useful for processing files from the GAIA API.
+    
+    Args:
+        content: The content to save to the file
+        filename: Optional filename, will generate a random name if not provided
+        
+    Returns:
+        Path to the saved file
+    """
+    temp_dir = tempfile.gettempdir()
+    if filename is None:
+        temp_file = tempfile.NamedTemporaryFile(delete=False)
+        filepath = temp_file.name
+    else:
+        filepath = os.path.join(temp_dir, filename)
+    
+    # Write content to the file
+    with open(filepath, 'w') as f:
+        f.write(content)
+    
+    return f"File saved to {filepath}. You can read this file to process its contents."
+
+@tool
+def download_file_from_url(url: str, filename: Optional[str] = None) -> str:
+    """
+    Download a file from a URL and save it to a temporary location.
+    
+    Args:
+        url: The URL to download from
+        filename: Optional filename, will generate one based on URL if not provided
+        
+    Returns:
+        Path to the downloaded file
+    """
+    try:
+        # Parse URL to get filename if not provided
+        if not filename:
+            path = urlparse(url).path
+            filename = os.path.basename(path)
+            if not filename:
+                # Generate a random name if we couldn't extract one
+                import uuid
+                filename = f"downloaded_{uuid.uuid4().hex[:8]}"
+        
+        # Create temporary file
+        temp_dir = tempfile.gettempdir()
+        filepath = os.path.join(temp_dir, filename)
+        
+        # Download the file
+        response = requests.get(url, stream=True)
+        response.raise_for_status()
+        
+        # Save the file
+        with open(filepath, 'wb') as f:
+            for chunk in response.iter_content(chunk_size=8192):
+                f.write(chunk)
+        
+        return f"File downloaded to {filepath}. You can now process this file."
+    except Exception as e:
+        return f"Error downloading file: {str(e)}"
+
+@tool
+def extract_text_from_image(image_path: str) -> str:
+    """
+    Extract text from an image using pytesseract (if available).
+    
+    Args:
+        image_path: Path to the image file
+        
+    Returns:
+        Extracted text or error message
+    """
+    try:
+        # Try to import pytesseract
+        import pytesseract
+        from PIL import Image
+        
+        # Open the image
+        image = Image.open(image_path)
+        
+        # Extract text
+        text = pytesseract.image_to_string(image)
+        
+        return f"Extracted text from image:\n\n{text}"
+    except ImportError:
+        return "Error: pytesseract is not installed. Please install it with 'pip install pytesseract' and ensure Tesseract OCR is installed on your system."
+    except Exception as e:
+        return f"Error extracting text from image: {str(e)}"
+
+@tool
+def analyze_csv_file(file_path: str, query: str) -> str:
+    """
+    Analyze a CSV file using pandas and answer a question about it.
+    
+    Args:
+        file_path: Path to the CSV file
+        query: Question about the data
+        
+    Returns:
+        Analysis result or error message
+    """
+    try:
+        import pandas as pd
+        
+        # Read the CSV file
+        df = pd.read_csv(file_path)
+        
+        # Run various analyses based on the query
+        result = f"CSV file loaded with {len(df)} rows and {len(df.columns)} columns.\n"
+        result += f"Columns: {', '.join(df.columns)}\n\n"
+        
+        # Add summary statistics
+        result += "Summary statistics:\n"
+        result += str(df.describe())
+        
+        return result
+    except ImportError:
+        return "Error: pandas is not installed. Please install it with 'pip install pandas'."
+    except Exception as e:
+        return f"Error analyzing CSV file: {str(e)}"
+
+@tool
+def analyze_excel_file(file_path: str, query: str) -> str:
+    """
+    Analyze an Excel file using pandas and answer a question about it.
+    
+    Args:
+        file_path: Path to the Excel file
+        query: Question about the data
+        
+    Returns:
+        Analysis result or error message
+    """
+    try:
+        import pandas as pd
+        
+        # Read the Excel file
+        df = pd.read_excel(file_path)
+        
+        # Run various analyses based on the query
+        result = f"Excel file loaded with {len(df)} rows and {len(df.columns)} columns.\n"
+        result += f"Columns: {', '.join(df.columns)}\n\n"
+        
+        # Add summary statistics
+        result += "Summary statistics:\n"
+        result += str(df.describe())
+        
+        return result
+    except ImportError:
+        return "Error: pandas and openpyxl are not installed. Please install them with 'pip install pandas openpyxl'."
+    except Exception as e:
+        return f"Error analyzing Excel file: {str(e)}"
+
+class GAIAAgent:
+    def __init__(
+        self, 
+        model_type: str = "HfApiModel", 
+        model_id: Optional[str] = None,
+        api_key: Optional[str] = None,
+        api_base: Optional[str] = None,
+        temperature: float = 0.2,
+        executor_type: str = "local",  # Changed from use_e2b to executor_type
+        additional_imports: List[str] = None,
+        additional_tools: List[Any] = None,
+        system_prompt: Optional[str] = None,  # We'll still accept this parameter but not use it directly
+        verbose: bool = False,
+        provider: Optional[str] = None,  # Add provider for InferenceClientModel
+        timeout: Optional[int] = None   # Add timeout for InferenceClientModel
+    ):
+        """
+        Initialize a GAIAAgent with specified configuration
+        
+        Args:
+            model_type: Type of model to use (HfApiModel, LiteLLMModel, OpenAIServerModel, InferenceClientModel)
+            model_id: ID of the model to use
+            api_key: API key for the model provider
+            api_base: Base URL for API calls
+            temperature: Temperature for text generation
+            executor_type: Type of executor for code execution ('local' or 'e2b')
+            additional_imports: Additional Python modules to allow importing
+            additional_tools: Additional tools to provide to the agent
+            system_prompt: Custom system prompt to use (not directly used, kept for backward compatibility)
+            verbose: Enable verbose logging
+            provider: Provider for InferenceClientModel (e.g., "hf-inference")
+            timeout: Timeout in seconds for API calls
+        """
+        # Set verbosity
+        self.verbose = verbose
+        self.system_prompt = system_prompt  # Store for potential future use
+        
+        # Initialize model based on configuration
+        if model_type == "HfApiModel":
+            if api_key is None:
+                api_key = os.getenv("HUGGINGFACEHUB_API_TOKEN")
+                if not api_key:
+                    raise ValueError("No Hugging Face token provided. Please set HUGGINGFACEHUB_API_TOKEN environment variable or pass api_key parameter.")
+            
+            if self.verbose:
+                print(f"Using Hugging Face token: {api_key[:5]}...")
+                
+            self.model = HfApiModel(
+                model_id=model_id or "meta-llama/Llama-3-70B-Instruct",
+                token=api_key,
+                temperature=temperature
+            )
+        elif model_type == "InferenceClientModel":
+            if api_key is None:
+                api_key = os.getenv("HUGGINGFACEHUB_API_TOKEN")
+                if not api_key:
+                    raise ValueError("No Hugging Face token provided. Please set HUGGINGFACEHUB_API_TOKEN environment variable or pass api_key parameter.")
+            
+            if self.verbose:
+                print(f"Using Hugging Face token: {api_key[:5]}...")
+                
+            self.model = InferenceClientModel(
+                model_id=model_id or "meta-llama/Llama-3-70B-Instruct",
+                provider=provider or "hf-inference",
+                token=api_key,
+                timeout=timeout or 120,
+                temperature=temperature
+            )
+        elif model_type == "LiteLLMModel":
+            from smolagents import LiteLLMModel
+            self.model = LiteLLMModel(
+                model_id=model_id or "gpt-4o",
+                api_key=api_key or os.getenv("OPENAI_API_KEY"),
+                temperature=temperature
+            )
+        elif model_type == "OpenAIServerModel":
+            # Check for xAI API key and base URL first
+            xai_api_key = os.getenv("XAI_API_KEY")
+            xai_api_base = os.getenv("XAI_API_BASE")
+            
+            # If xAI credentials are available, use them
+            if xai_api_key and api_key is None:
+                api_key = xai_api_key
+                if self.verbose:
+                    print(f"Using xAI API key: {api_key[:5]}...")
+            
+            # If no API key specified, fall back to OPENAI_API_KEY
+            if api_key is None:
+                api_key = os.getenv("OPENAI_API_KEY")
+                if not api_key:
+                    raise ValueError("No OpenAI API key provided. Please set OPENAI_API_KEY or XAI_API_KEY environment variable or pass api_key parameter.")
+            
+            # If xAI API base is available and no api_base is provided, use it
+            if xai_api_base and api_base is None:
+                api_base = xai_api_base
+                if self.verbose:
+                    print(f"Using xAI API base URL: {api_base}")
+            
+            # If no API base specified but environment variable available, use it
+            if api_base is None:
+                api_base = os.getenv("AGENT_API_BASE")
+                if api_base and self.verbose:
+                    print(f"Using API base from AGENT_API_BASE: {api_base}")
+            
+            self.model = OpenAIServerModel(
+                model_id=model_id or "gpt-4o",
+                api_key=api_key,
+                api_base=api_base,
+                temperature=temperature
+            )
+        else:
+            raise ValueError(f"Unknown model type: {model_type}")
+        
+        if self.verbose:
+            print(f"Initialized model: {model_type} - {model_id}")
+        
+        # Initialize default tools
+        self.tools = [
+            DuckDuckGoSearchTool(),
+            PythonInterpreterTool(),
+            save_and_read_file,
+            download_file_from_url,
+            analyze_csv_file,
+            analyze_excel_file
+        ]
+        
+        # Add extract_text_from_image if PIL and pytesseract are available
+        try:
+            import pytesseract
+            from PIL import Image
+            self.tools.append(extract_text_from_image)
+            if self.verbose:
+                print("Added image processing tool")
+        except ImportError:
+            if self.verbose:
+                print("Image processing libraries not available")
+        
+        # Add any additional tools
+        if additional_tools:
+            self.tools.extend(additional_tools)
+            
+        if self.verbose:
+            print(f"Initialized with {len(self.tools)} tools")
+        
+        # Setup imports allowed
+        self.imports = ["pandas", "numpy", "datetime", "json", "re", "math", "os", "requests", "csv", "urllib"]
+        if additional_imports:
+            self.imports.extend(additional_imports)
+            
+        # Initialize the CodeAgent
+        executor_kwargs = {}
+        if executor_type == "e2b":
+            try:
+                # Try to import e2b dependencies to check if they're available
+                from e2b_code_interpreter import Sandbox
+                if self.verbose:
+                    print("Using e2b executor")
+            except ImportError:
+                if self.verbose:
+                    print("e2b dependencies not found, falling back to local executor")
+                executor_type = "local"  # Fallback to local if e2b is not available
+        
+        self.agent = CodeAgent(
+            tools=self.tools,
+            model=self.model,
+            additional_authorized_imports=self.imports,
+            executor_type=executor_type,
+            executor_kwargs=executor_kwargs,
+            verbosity_level=2 if self.verbose else 0
+        )
+        
+        if self.verbose:
+            print("Agent initialized and ready")
+    
+    def answer_question(self, question: str, task_file_path: Optional[str] = None) -> str:
+        """
+        Process a GAIA benchmark question and return the answer
+        
+        Args:
+            question: The question to answer
+            task_file_path: Optional path to a file associated with the question
+            
+        Returns:
+            The answer to the question
+        """
+        try:
+            if self.verbose:
+                print(f"Processing question: {question}")
+                if task_file_path:
+                    print(f"With associated file: {task_file_path}")
+            
+            # Create a context with file information if available
+            context = question
+            file_content = None
+            
+            # If there's a file, read it and include its content in the context
+            if task_file_path:
+                try:
+                    with open(task_file_path, 'r') as f:
+                        file_content = f.read()
+                    
+                    # Determine file type from extension
+                    import os
+                    file_ext = os.path.splitext(task_file_path)[1].lower()
+                    
+                    context = f"""
+Question: {question}
+
+This question has an associated file. Here is the file content:
+
+```{file_ext}
+{file_content}
+```
+
+Analyze the file content above to answer the question.
+"""
+                except Exception as file_e:
+                    context = f"""
+Question: {question}
+
+This question has an associated file at path: {task_file_path}
+However, there was an error reading the file: {file_e}
+You can still try to answer the question based on the information provided.
+"""
+            
+            # Check for special cases that need specific formatting
+            # Reversed text questions
+            if question.startswith(".") or ".rewsna eht sa" in question:
+                context = f"""
+This question appears to be in reversed text. Here's the reversed version:
+{question[::-1]}
+
+Now answer the question above. Remember to format your answer exactly as requested.
+"""
+            
+            # Add a prompt to ensure precise answers
+            full_prompt = f"""{context}
+
+When answering, provide ONLY the precise answer requested. 
+Do not include explanations, steps, reasoning, or additional text.
+Be direct and specific. GAIA benchmark requires exact matching answers.
+For example, if asked "What is the capital of France?", respond simply with "Paris".
+"""
+            
+            # Run the agent with the question
+            answer = self.agent.run(full_prompt)
+            
+            # Clean up the answer to ensure it's in the expected format
+            # Remove common prefixes that models often add
+            answer = self._clean_answer(answer)
+            
+            if self.verbose:
+                print(f"Generated answer: {answer}")
+                
+            return answer
+        except Exception as e:
+            error_msg = f"Error answering question: {e}"
+            if self.verbose:
+                print(error_msg)
+            return error_msg
+    
+    def _clean_answer(self, answer: any) -> str:
+        """
+        Clean up the answer to remove common prefixes and formatting
+        that models often add but that can cause exact match failures.
+        
+        Args:
+            answer: The raw answer from the model
+            
+        Returns:
+            The cleaned answer as a string
+        """
+        # Convert non-string types to strings
+        if not isinstance(answer, str):
+            # Handle numeric types (float, int)
+            if isinstance(answer, float):
+                # Format floating point numbers properly
+                # Check if it's an integer value in float form (e.g., 12.0)
+                if answer.is_integer():
+                    formatted_answer = str(int(answer))
+                else:
+                    # For currency values that might need formatting
+                    if abs(answer) >= 1000:
+                        formatted_answer = f"${answer:,.2f}"
+                    else:
+                        formatted_answer = str(answer)
+                return formatted_answer
+            elif isinstance(answer, int):
+                return str(answer)
+            else:
+                # For any other type
+                return str(answer)
+        
+        # Now we know answer is a string, so we can safely use string methods
+        # Normalize whitespace
+        answer = answer.strip()
+        
+        # Remove common prefixes and formatting that models add
+        prefixes_to_remove = [
+            "The answer is ", 
+            "Answer: ",
+            "Final answer: ",
+            "The result is ",
+            "To answer this question: ",
+            "Based on the information provided, ",
+            "According to the information: ",
+        ]
+        
+        for prefix in prefixes_to_remove:
+            if answer.startswith(prefix):
+                answer = answer[len(prefix):].strip()
+        
+        # Remove quotes if they wrap the entire answer
+        if (answer.startswith('"') and answer.endswith('"')) or (answer.startswith("'") and answer.endswith("'")):
+            answer = answer[1:-1].strip()
+        
+        return answer

local_test.py

@@ -0,0 +1,249 @@
+#!/usr/bin/env python3
+"""
+Test script for the GAIA agent using real API keys.
+This script simulates GAIA benchmark questions and helps debug/improve the agent.
+"""
+
+import os
+import sys
+import json
+import tempfile
+from typing import List, Dict, Any, Optional
+import traceback
+import dotenv
+
+# Load environment variables from .env file
+dotenv.load_dotenv()
+
+# Import our agent
+from core_agent import GAIAAgent
+
+# Simulation of GAIA benchmark questions
+SAMPLE_QUESTIONS = [
+    {
+        "task_id": "task_001",
+        "question": "What is the capital of France?",
+        "expected_answer": "Paris",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_002",
+        "question": "What is the square root of 144?",
+        "expected_answer": "12",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_003",
+        "question": "If a train travels at 60 miles per hour, how far will it travel in 2.5 hours?",
+        "expected_answer": "150 miles",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_004", 
+        "question": ".rewsna eht sa 'thgir' drow eht etirw ,tfel fo etisoppo eht si tahW",
+        "expected_answer": "right",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_005",
+        "question": "Analyze the data in the attached CSV file and tell me the total sales for the month of January.",
+        "expected_answer": "$10,250.75",
+        "has_file": True,
+        "file_content": """Date,Product,Quantity,Price,Total
+2023-01-05,Widget A,10,25.99,259.90
+2023-01-12,Widget B,5,45.50,227.50
+2023-01-15,Widget C,20,50.25,1005.00
+2023-01-20,Widget A,15,25.99,389.85
+2023-01-25,Widget B,8,45.50,364.00
+2023-01-28,Widget D,100,80.04,8004.50"""
+    },
+    {
+        "task_id": "task_006",
+        "question": "I'm making a grocery list for my mom, but she's a picky eater. She only eats foods that don't contain the letter 'e'. List 5 common fruits and vegetables she can eat.",
+        "expected_answer": "Banana, Kiwi, Corn, Fig, Taro",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_007",
+        "question": "How many studio albums were published by Mercedes Sosa between 1972 and 1985?",
+        "expected_answer": "12",
+        "has_file": False,
+        "file_content": None
+    },
+    {
+        "task_id": "task_008",
+        "question": "In the video https://www.youtube.com/watch?v=L1vXC1KMRd0, what color is primarily associated with the main character?",
+        "expected_answer": "Blue",
+        "has_file": False,
+        "file_content": None
+    }
+]
+
+def initialize_agent():
+    """Initialize the GAIAAgent with appropriate API keys."""
+    print("Initializing GAIAAgent with API keys...")
+    
+    # Try X.AI first (xAI) with the correct API endpoint
+    if os.getenv("XAI_API_KEY"):
+        print("Using X.AI API key")
+        try:
+            agent = GAIAAgent(
+                model_type="OpenAIServerModel",
+                model_id="grok-3-latest",  # Use the X.AI model
+                api_key=os.getenv("XAI_API_KEY"),
+                api_base="https://api.x.ai/v1",  # Correct X.AI endpoint
+                temperature=0.2,
+                executor_type="local",
+                verbose=True,
+                system_prompt_suffix=additional_system_prompt  # Add our hints
+            )
+            print("Using OpenAIServerModel with X.AI API")
+            return agent
+        except Exception as e:
+            print(f"Error initializing with X.AI API: {e}")
+            traceback.print_exc()
+    
+    # Then try OpenAI
+    if os.getenv("OPENAI_API_KEY"):
+        print("Using OpenAI API key")
+        try:
+            model_id = os.getenv("AGENT_MODEL_ID", "gpt-4o")
+            agent = GAIAAgent(
+                model_type="OpenAIServerModel",
+                model_id=model_id,
+                api_key=os.getenv("OPENAI_API_KEY"),
+                temperature=0.2,
+                executor_type="local",
+                verbose=True
+            )
+            print(f"Using OpenAIServerModel with model_id: {model_id}")
+            return agent
+        except Exception as e:
+            print(f"Error initializing with OpenAI API: {e}")
+            traceback.print_exc()
+    
+    # Last resort, try Hugging Face
+    if os.getenv("HUGGINGFACEHUB_API_TOKEN"):
+        print("Using Hugging Face API token")
+        try:
+            # Use a smaller model that might work within free tier
+            model_id = "tiiuae/falcon-7b-instruct"  # Try a smaller model that might be within free tier
+            agent = GAIAAgent(
+                model_type="HfApiModel",
+                model_id=model_id,
+                api_key=os.getenv("HUGGINGFACEHUB_API_TOKEN"),
+                temperature=0.2,
+                executor_type="local",
+                verbose=True
+            )
+            print(f"Using HfApiModel with model_id: {model_id}")
+            return agent
+        except Exception as e:
+            print(f"Error initializing with Hugging Face API: {e}")
+            traceback.print_exc()
+    
+    print("ERROR: No valid API keys found in environment. Please set one of the following:")
+    print("- XAI_API_KEY (for X.AI)")
+    print("- OPENAI_API_KEY")
+    print("- HUGGINGFACEHUB_API_TOKEN")
+    return None
+
+def save_test_file(task_id: str, content: str) -> str:
+    """Save a test file to a temporary location."""
+    temp_dir = tempfile.gettempdir()
+    file_path = os.path.join(temp_dir, f"test_file_{task_id}.csv")
+    
+    with open(file_path, 'w') as f:
+        f.write(content)
+    
+    return file_path
+
+def run_tests():
+    """Run tests using the GAIAAgent with API keys."""
+    agent = initialize_agent()
+    
+    if not agent:
+        print("Failed to initialize agent. Exiting.")
+        return
+    
+    results = []
+    correct_count = 0
+    total_count = len(SAMPLE_QUESTIONS)
+    
+    for idx, question_data in enumerate(SAMPLE_QUESTIONS):
+        task_id = question_data["task_id"]
+        question = question_data["question"]
+        expected = question_data["expected_answer"]
+        
+        print(f"\n{'='*80}")
+        print(f"Question {idx+1}/{total_count}: {question}")
+        print(f"Expected: {expected}")
+        
+        # Process any attached file
+        file_path = None
+        if question_data["has_file"] and question_data["file_content"]:
+            file_path = save_test_file(task_id, question_data["file_content"])
+            print(f"Created test file: {file_path}")
+        
+        # Get answer from agent
+        try:
+            answer = agent.answer_question(question, file_path)
+            print(f"Agent answer: {answer}")
+            
+            # Check if answer matches expected
+            is_correct = answer.lower() == expected.lower()
+            if is_correct:
+                correct_count += 1
+                print(f"✅ CORRECT")
+            else:
+                print(f"❌ INCORRECT - Expected: {expected}")
+            
+            results.append({
+                "task_id": task_id,
+                "question": question,
+                "expected": expected,
+                "answer": answer,
+                "is_correct": is_correct
+            })
+        except Exception as e:
+            error_details = traceback.format_exc()
+            print(f"Error processing question: {e}\n{error_details}")
+            results.append({
+                "task_id": task_id,
+                "question": question,
+                "expected": expected,
+                "answer": f"ERROR: {str(e)}",
+                "is_correct": False
+            })
+    
+    # Print summary
+    accuracy = (correct_count / total_count) * 100
+    print(f"\n{'='*80}")
+    print(f"Test Results: {correct_count}/{total_count} correct ({accuracy:.1f}%)")
+    
+    return results
+
+
+if __name__ == "__main__":
+    print("Running tests for GAIA agent with API keys...")
+    
+    # Print environment information
+    print("\nEnvironment information:")
+    print(f"XAI_API_KEY set: {'Yes' if os.getenv('XAI_API_KEY') else 'No'}")
+    print(f"OPENAI_API_KEY set: {'Yes' if os.getenv('OPENAI_API_KEY') else 'No'}")
+    print(f"HUGGINGFACEHUB_API_TOKEN set: {'Yes' if os.getenv('HUGGINGFACEHUB_API_TOKEN') else 'No'}")
+    print(f"AGENT_MODEL_TYPE: {os.getenv('AGENT_MODEL_TYPE', 'OpenAIServerModel')} (default: OpenAIServerModel)")
+    print(f"AGENT_MODEL_ID: {os.getenv('AGENT_MODEL_ID', 'gpt-4o')} (default: gpt-4o)")
+    
+    results = run_tests()
+    
+    # Save results to a file
+    with open("test_results.json", "w") as f:
+        json.dump(results, f, indent=2)
+    
+    print("\nResults saved to test_results.json") 

main.py

@@ -0,0 +1,277 @@
+import os
+import tempfile
+import gradio as gr
+import pandas as pd
+import traceback
+from core_agent import GAIAAgent
+from api_integration import GAIAApiClient
+
+# Constants
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+def save_task_file(file_content, task_id):
+    """
+    Save a task file to a temporary location
+    """
+    if not file_content:
+        return None
+    
+    # Create a temporary file
+    temp_dir = tempfile.gettempdir()
+    file_path = os.path.join(temp_dir, f"gaia_task_{task_id}.txt")
+    
+    # Write content to the file
+    with open(file_path, 'wb') as f:
+        f.write(file_content)
+    
+    print(f"File saved to {file_path}")
+    return file_path
+
+def get_agent_configuration():
+    """
+    Get the agent configuration based on environment variables
+    """
+    # Default configuration
+    config = {
+        "model_type": "OpenAIServerModel",  # Default to OpenAIServerModel
+        "model_id": "gpt-4o",  # Default model for OpenAI
+        "temperature": 0.2,
+        "executor_type": "local",
+        "verbose": False,
+        "provider": "hf-inference",  # For InferenceClientModel
+        "timeout": 120        # For InferenceClientModel
+    }
+    
+    # Check for xAI API key and base URL
+    xai_api_key = os.getenv("XAI_API_KEY")
+    xai_api_base = os.getenv("XAI_API_BASE")
+    
+    # If we have xAI credentials, use them
+    if xai_api_key:
+        config["api_key"] = xai_api_key
+        if xai_api_base:
+            config["api_base"] = xai_api_base
+            # Use a model that works well with xAI
+            config["model_id"] = "mixtral-8x7b-32768"
+    
+    # Override with environment variables if present
+    if os.getenv("AGENT_MODEL_TYPE"):
+        config["model_type"] = os.getenv("AGENT_MODEL_TYPE")
+    
+    if os.getenv("AGENT_MODEL_ID"):
+        config["model_id"] = os.getenv("AGENT_MODEL_ID")
+    
+    if os.getenv("AGENT_TEMPERATURE"):
+        config["temperature"] = float(os.getenv("AGENT_TEMPERATURE"))
+    
+    if os.getenv("AGENT_EXECUTOR_TYPE"):
+        config["executor_type"] = os.getenv("AGENT_EXECUTOR_TYPE")
+    
+    if os.getenv("AGENT_VERBOSE") is not None:
+        config["verbose"] = os.getenv("AGENT_VERBOSE").lower() == "true"
+    
+    if os.getenv("AGENT_API_BASE"):
+        config["api_base"] = os.getenv("AGENT_API_BASE")
+    
+    # InferenceClientModel specific settings
+    if os.getenv("AGENT_PROVIDER"):
+        config["provider"] = os.getenv("AGENT_PROVIDER")
+    
+    if os.getenv("AGENT_TIMEOUT"):
+        config["timeout"] = int(os.getenv("AGENT_TIMEOUT"))
+    
+    return config
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    """
+    Fetches all questions, runs the GAIAAgent on them, submits all answers,
+    and displays the results.
+    """
+    # Check for user login
+    if not profile:
+        return "Please Login to Hugging Face with the button.", None
+    
+    username = profile.username
+    print(f"User logged in: {username}")
+    
+    # Get SPACE_ID for code link
+    space_id = os.getenv("SPACE_ID")
+    agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
+    
+    # Initialize API client
+    api_client = GAIAApiClient(DEFAULT_API_URL)
+    
+    # Initialize Agent with configuration
+    try:
+        agent_config = get_agent_configuration()
+        print(f"Using agent configuration: {agent_config}")
+        
+        agent = GAIAAgent(**agent_config)
+        print("Agent initialized successfully")
+    except Exception as e:
+        error_details = traceback.format_exc()
+        print(f"Error initializing agent: {e}\n{error_details}")
+        return f"Error initializing agent: {e}", None
+    
+    # Fetch questions
+    try:
+        questions_data = api_client.get_questions()
+        if not questions_data:
+            return "Fetched questions list is empty or invalid format.", None
+        print(f"Fetched {len(questions_data)} questions.")
+    except Exception as e:
+        error_details = traceback.format_exc()
+        print(f"Error fetching questions: {e}\n{error_details}")
+        return f"Error fetching questions: {e}", None
+    
+    # Run agent on questions
+    results_log = []
+    answers_payload = []
+    print(f"Running agent on {len(questions_data)} questions...")
+    
+    # Progress tracking
+    total_questions = len(questions_data)
+    completed = 0
+    failed = 0
+    
+    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:
+            # Update progress
+            completed += 1
+            print(f"Processing question {completed}/{total_questions}: Task ID {task_id}")
+            
+            # Check if the question has an associated file
+            file_path = None
+            try:
+                file_content = api_client.get_file(task_id)
+                print(f"Downloaded file for task {task_id}")
+                file_path = save_task_file(file_content, task_id)
+            except Exception as file_e:
+                print(f"No file found for task {task_id} or error: {file_e}")
+            
+            # Run the agent to get the answer
+            submitted_answer = agent.answer_question(question_text, file_path)
+            
+            # Add to results
+            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:
+            # Update error count
+            failed += 1
+            error_details = traceback.format_exc()
+            print(f"Error running agent on task {task_id}: {e}\n{error_details}")
+            
+            # Add error to results
+            error_msg = f"AGENT ERROR: {e}"
+            answers_payload.append({"task_id": task_id, "submitted_answer": error_msg})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text,
+                "Submitted Answer": error_msg
+            })
+    
+    # Print summary
+    print(f"\nProcessing complete: {completed} questions processed, {failed} failures")
+    
+    if not answers_payload:
+        return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
+    
+    # Submit answers
+    submission_data = {
+        "username": username.strip(),
+        "agent_code": agent_code,
+        "answers": answers_payload
+    }
+    
+    print(f"Submitting {len(answers_payload)} answers for username '{username}'...")
+    
+    try:
+        result_data = api_client.submit_answers(
+            username.strip(),
+            agent_code,
+            answers_payload
+        )
+        
+        # Calculate success rate
+        correct_count = result_data.get('correct_count', 0)
+        total_attempted = result_data.get('total_attempted', len(answers_payload))
+        success_rate = (correct_count / total_attempted) * 100 if total_attempted > 0 else 0
+        
+        final_status = (
+            f"Submission Successful!\n"
+            f"User: {result_data.get('username')}\n"
+            f"Overall Score: {result_data.get('score', 'N/A')}% "
+            f"({correct_count}/{total_attempted} correct, {success_rate:.1f}% success rate)\n"
+            f"Message: {result_data.get('message', 'No message received.')}"
+        )
+        
+        print("Submission successful.")
+        return final_status, pd.DataFrame(results_log)
+    except Exception as e:
+        error_details = traceback.format_exc()
+        status_message = f"Submission Failed: {e}\n{error_details}"
+        print(status_message)
+        return status_message, pd.DataFrame(results_log)
+
+# Build Gradio Interface
+with gr.Blocks() as demo:
+    gr.Markdown("# GAIA Agent Evaluation Runner")
+    gr.Markdown(
+        """
+        **Instructions:**
+        
+        1. Log in to your Hugging Face account using the button below.
+        2. Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit answers, and see the score.
+        
+        **Configuration:**
+        
+        You can configure the agent by setting these environment variables:
+        - `AGENT_MODEL_TYPE`: Model type (HfApiModel, InferenceClientModel, LiteLLMModel, OpenAIServerModel)
+        - `AGENT_MODEL_ID`: Model ID
+        - `AGENT_TEMPERATURE`: Temperature for generation (0.0-1.0)
+        - `AGENT_EXECUTOR_TYPE`: Type of executor ('local' or 'e2b')
+        - `AGENT_VERBOSE`: Enable verbose logging (true/false)
+        - `AGENT_API_BASE`: Base URL for API calls (for OpenAIServerModel)
+        
+        **xAI Support:**
+        - `XAI_API_KEY`: Your xAI API key
+        - `XAI_API_BASE`: Base URL for xAI API (default: https://api.groq.com/openai/v1)
+        - When using xAI, set AGENT_MODEL_TYPE=OpenAIServerModel and AGENT_MODEL_ID=mixtral-8x7b-32768
+        
+        **InferenceClientModel specific settings:**
+        - `AGENT_PROVIDER`: Provider for InferenceClientModel (e.g., "hf-inference") 
+        - `AGENT_TIMEOUT`: Timeout in seconds for API calls
+        """
+    )
+    
+    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)
+    
+    # Check for environment variables
+    config = get_agent_configuration()
+    print(f"Agent configuration: {config}")
+    
+    # Run the Gradio app
+    demo.launch(debug=True, share=False)

requirements.txt

@@ -0,0 +1,6 @@
+gradio
+requests
+smolagents
+python-dotenv
+pandas
+numpy