Update dependencies in pyproject.toml and requirements.txt, and add test_single_q.py for running single and all question tests. Introduce tools.py for Perplexity API integration with enhanced error handling and logging.

pyproject.toml

@@ -5,5 +5,13 @@ description = "Add your description here"
 readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [
+    "bs4>=0.0.2",
+    "dotenv>=0.9.9",
     "duckduckgo-search>=8.0.2",
+    "gradio>=5.29.1",
+    "markdownify>=1.1.0",
+    "openai>=1.75.0",
+    "openpyxl>=3.1.5",
+    "requests>=2.32.3",
+    "smolagents[litellm]>=1.16.1",
 ]

requirements.txt

@@ -1,2 +1,5 @@
 gradio
-requests
+requests
+openai
+smolagents
+openpyxl

test_single_q.py

@@ -0,0 +1,149 @@
+import json
+import os
+import requests
+import sys # Added for command-line arguments
+from agent import BasicAgent
+
+# --- Constants ---
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+DOWNLOAD_DIR = "downloaded_task_files" # Directory to save downloaded files
+QUESTIONS_FILE = "data/questions.json"
+# QUESTION_INDEX = 3 # Removed global constant
+
+def run_single_question_test(question_index: int): # Added question_index parameter
+    """
+    Fetches a single question by index, downloads its associated file (if any),
+    runs the BasicAgent on it, and prints the answer.
+    """
+    # Create download directory if it doesn't exist
+    try:
+        os.makedirs(DOWNLOAD_DIR, exist_ok=True)
+        print(f"Ensured download directory exists: {DOWNLOAD_DIR}")
+    except OSError as e:
+        print(f"Error creating download directory {DOWNLOAD_DIR}: {e}")
+        return
+
+    # 1. Load Questions
+    try:
+        with open(QUESTIONS_FILE, 'r') as f:
+            questions_data = json.load(f)
+        if not questions_data or not isinstance(questions_data, list) or len(questions_data) <= question_index:
+            print(f"Error: Could not load question at index {question_index} (0-indexed) from {QUESTIONS_FILE}. Total questions: {len(questions_data) if isinstance(questions_data, list) else 'N/A'}")
+            return
+        item = questions_data[question_index]
+        print(f"Loaded question {question_index + 1} (from 1-indexed input): {item.get('question')}")
+    except FileNotFoundError:
+        print(f"Error: Questions file not found at {QUESTIONS_FILE}")
+        return
+    except json.JSONDecodeError:
+        print(f"Error: Could not decode JSON from {QUESTIONS_FILE}")
+        return
+    except Exception as e:
+        print(f"Error loading or parsing {QUESTIONS_FILE}: {e}")
+        return
+
+    # 2. Instantiate Agent
+    try:
+        agent = BasicAgent()
+        print("Agent instantiated successfully.")
+    except Exception as e:
+        print(f"Error instantiating agent: {e}")
+        return
+
+    task_id = item.get("task_id")
+    question_text = item.get("question")
+    file_name = item.get("file_name")
+    local_file_path = None
+
+    if not task_id or question_text is None:
+        print(f"Skipping item with missing task_id or question: {item}")
+        return
+
+    if file_name:
+        download_url = f"{DEFAULT_API_URL}/files/{task_id}"
+        local_file_path = os.path.join(DOWNLOAD_DIR, file_name)
+        print(f"Attempting to download file for task {task_id} (using task_id in URL): {file_name} from {download_url}")
+        try:
+            file_response = requests.get(download_url, stream=True, timeout=30)
+            file_response.raise_for_status()
+            with open(local_file_path, 'wb') as f:
+                for chunk in file_response.iter_content(chunk_size=8192):
+                    f.write(chunk)
+            print(f"Successfully downloaded to {local_file_path}")
+        except requests.exceptions.RequestException as e:
+            print(f"Failed to download {file_name}: {e}")
+            local_file_path = None # Ensure agent doesn't get a path to a non-existent/failed file
+        except Exception as e:
+            print(f"An unexpected error occurred downloading {file_name}: {e}")
+            local_file_path = None
+    else:
+        print("No file associated with this question.")
+
+    # 3. Run Agent on the single question
+    try:
+        print(f"Running agent on question: {question_text}")
+        if local_file_path:
+            print(f"Providing file: {local_file_path}")
+        submitted_answer = agent(question_text, file_path=local_file_path)
+        print(f"\n--- Agent's Answer for Task ID: {task_id} ---")
+        print(submitted_answer)
+        print("--- End of Answer ---")
+
+    except Exception as e:
+         print(f"Error running agent on task {task_id}: {e}")
+
+def run_all_questions_test():
+    """
+    Runs the test for all questions in the QUESTIONS_FILE.
+    """
+    print("Attempting to run tests for all questions...")
+    try:
+        with open(QUESTIONS_FILE, 'r') as f:
+            questions_data = json.load(f)
+        if not questions_data or not isinstance(questions_data, list):
+            print(f"Error: Could not load questions from {QUESTIONS_FILE} or it's not a list.")
+            return
+        
+        num_questions = len(questions_data)
+        if num_questions == 0:
+            print(f"No questions found in {QUESTIONS_FILE}.")
+            return
+
+        print(f"Found {num_questions} questions to test.")
+        for i in range(num_questions):
+            print(f"\n--- Running Test for Question {i+1}/{num_questions} ---")
+            run_single_question_test(i)
+            print(f"--- Finished Test for Question {i+1}/{num_questions} ---\n")
+        print("All question tests completed.")
+
+    except FileNotFoundError:
+        print(f"Error: Questions file not found at {QUESTIONS_FILE}")
+    except json.JSONDecodeError:
+        print(f"Error: Could not decode JSON from {QUESTIONS_FILE}")
+    except Exception as e:
+        print(f"An unexpected error occurred while running all questions: {e}")
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("Usage: python test_single_q.py <question_number | 'all'>")
+        print("Example (single question): python test_single_q.py 4")
+        print("Example (all questions): python test_single_q.py all")
+        run_all_questions_test() # Default to running all if no argument is provided
+        sys.exit(0)
+
+    argument = sys.argv[1]
+
+    if argument.lower() == "all":
+        run_all_questions_test()
+    else:
+        try:
+            question_number_arg = int(argument)
+            if question_number_arg <= 0:
+                print("Error: Question number must be a positive integer.")
+                sys.exit(1)
+            # Convert 1-indexed input to 0-indexed for list access
+            question_idx_0_based = question_number_arg - 1
+            run_single_question_test(question_idx_0_based)
+        except ValueError:
+            print("Error: Invalid argument. Please enter an integer for a specific question or 'all'.")
+            sys.exit(1)

tools.py

@@ -0,0 +1,68 @@
+from openai import OpenAI
+from smolagents import tool
+import os
+from dotenv import load_dotenv
+
+load_dotenv()
+
+PERPLEXITY_API_KEY = os.getenv("PERPLEXITY_API_KEY")
+
+
+@tool
+def perplexity_search_tool(query: str) -> str:
+    """
+    Performs a web search using the Perplexity API and returns the result.
+
+    Args:
+        query: The search query.
+
+    Returns:
+        The search result from Perplexity API or an error message.
+    """
+    if not PERPLEXITY_API_KEY:
+        return "Error: Perplexity API key not set. Please set the PERPLEXITY_API_KEY environment variable."
+
+    messages = [
+        {
+            "role": "system",
+            "content": (
+                "You are a helpful assistant"
+            ),
+        },
+        {
+            "role": "user",
+            "content": query,
+        },
+    ]
+
+    try:
+        client = OpenAI(api_key=PERPLEXITY_API_KEY, base_url="https://api.perplexity.ai")
+
+        response = client.chat.completions.create(
+            model="sonar-pro",  # Using sonar-pro as per documentation
+            messages=messages,
+        )
+        # Assuming the main content is in the first choice's message
+        if response.choices and len(response.choices) > 0:
+            return response.choices[0].message.content
+        else:
+            return "Error: No response choices received from Perplexity API."
+    except Exception as e:
+        return f"Error calling Perplexity API: {str(e)}"
+
+
+if __name__ == "__main__":
+    # Example usage:
+    # Make sure to set the PERPLEXITY_API_KEY environment variable before running directly
+    if PERPLEXITY_API_KEY:
+        search_query = "What are the latest advancements in AI?"
+        result = perplexity_search_tool(search_query)
+        print(f"Search query: {search_query}")
+        print(f"Result:\\n{result}")
+
+        search_query_stars = "How many stars are there in our galaxy?"
+        result_stars = perplexity_search_tool(search_query_stars)
+        print(f"Search query: {search_query_stars}")
+        print(f"Result:\\n{result_stars}")
+    else:
+        print("Please set the PERPLEXITY_API_KEY environment variable to test the tool.")