app.py

# import os
# import gradio as gr
# import requests
# import inspect
# import pandas as pd

# # (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.")
#     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):
#     """
#     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)

##################################
#
# =================================================================================================
#  ✅ --- ✅  FINAL ASSESSMENT AGENT - INSTRUCTOR'S VERSION ✅ --- ✅
# =================================================================================================
#
#  Instructions:
#  1. Make sure you have a requirements.txt file with all the necessary packages.
#  2. Set your GROQ_API_KEY in the Hugging Face Space secrets.
#  3. This code replaces the original template entirely.
#
# =================================================================================================
# =================================================================================================
#  ✅ --- ✅  FINAL ASSESSMENT AGENT - INSTRUCTOR'S CORRECTED VERSION ✅ --- ✅
# =================================================================================================
#
#  Instructions:
#  1. Make sure your requirements.txt file matches the one provided by the instructor.
#  2. Set your GROQ_API_KEY in the Hugging Face Space secrets.
#  3. This code replaces the original template entirely.
#
# =================================================================================================

# 

###########################
# =================================================================================================
#  ✅ --- ✅  FINAL ASSESSMENT AGENT - V4 (STATE-FIXED & TAVILY) ✅ --- ✅
# =================================================================================================
#
#  Instructions:
#  1. Add TAVILY_API_KEY and GROQ_API_KEY to your HF Space secrets.
#  2. Update your requirements.txt to include `tavily-python`.
#  3. This version fixes the critical state-leakage bug and uses a better search tool.
#
# =================================================================================================

# 

######################
# =================================================================================================
#  ✅ --- ✅  FINAL ASSESSMENT AGENT - V5 (GPT-4o & PDF Support) ✅ --- ✅
# =================================================================================================
#
#  Instructions:
#  1. Add OPENAI_API_KEY, TAVILY_API_KEY, and GROQ_API_KEY to your HF Space secrets.
#  2. Update your requirements.txt to include `langchain-openai` and `pypdf`.
#  3. This version uses the GPT-4o model for superior reasoning and can read PDFs.
#
# =================================================================================================

# 
import os
import io
import json
import requests
import pandas as pd
import gradio as gr
from contextlib import redirect_stdout
from typing import TypedDict, Annotated, List
import operator

# --- LangChain & LangGraph Imports ---
from langchain_core.messages import BaseMessage, HumanMessage, ToolMessage, AIMessage, SystemMessage
from langchain_core.tools import tool
from langchain_huggingface import HuggingFaceEndpoint
from langgraph.graph import StateGraph, END
from tavily import TavilyClient
import pypdf

# --- Constants ---
DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
FILES_DIR = "./files"
os.makedirs(FILES_DIR, exist_ok=True)

# --- System Prompt (Updated for Manual JSON Tool Calling) ---
# This prompt instructs the model to generate JSON, a robust method for tool calls.
AGENT_SYSTEM_PROMPT = """You are a world-class AI agent, specialized in solving complex problems from the GAIA benchmark.
Your task is to analyze the user's question, think step-by-step, and use the provided tools to find the correct answer.

**TOOL USAGE INSTRUCTIONS:**
When you need to use a tool, you MUST respond with a JSON object containing the tool name and its arguments. The JSON object should have two keys: "tool_name" and "parameters".

Here is an example of how to call the `tavily_search` tool:
```json
{
  "tool_name": "tavily_search",
  "parameters": {
    "query": "Who won the last FIFA World Cup?"
  }
}
Use code with caution.
Python
CRITICAL FINAL ANSWER INSTRUCTIONS:
Once you have gathered all the necessary information and are absolutely certain of the answer, you MUST provide it directly and concisely.
Your final response must ONLY be the answer itself.
DO NOT wrap the final answer in a JSON object or include any conversational text.
Think, use your tools, and then provide ONLY the final, precise answer.
"""
###===============================================================================================
tavily = TavilyClient(api_key=os.getenv("TAVILY_API_KEY"))
@tool
def tavily_search(query: str) -> str:
"""Uses the Tavily Search API to find information on the web."""
print(f"--- Calling Tavily Search Tool with query: {query} ---")
try:
result = tavily.search(query=query, search_depth="advanced")
return f"Search results for '{query}':\n" + "\n".join([f"- {r['content']}" for r in result['results']])
except Exception as e: return f"Error during Tavily search: {e}"
@tool
def read_file(url: str) -> str:
"""Downloads and reads the content of a file (text or PDF) from a URL."""
print(f"--- Calling Read File Tool with URL: {url} ---")
try:
filename = os.path.join(FILES_DIR, os.path.basename(url))
response = requests.get(url)
response.raise_for_status()
with open(filename, 'wb') as f: f.write(response.content)
if url.lower().endswith('.pdf'):
try:
pdf_reader = pypdf.PdfReader(filename)
return f"Successfully read PDF file '{filename}'. Content:\n\n{''.join(p.extract_text() for p in pdf_reader.pages)}"
except Exception as e: return f"Error reading PDF file: {e}"
else:
try:
with open(filename, 'r', encoding='utf-8') as f: return f"Successfully read text file '{filename}'. Content:\n\n{f.read()}"
except UnicodeDecodeError: return f"Successfully downloaded binary file '{filename}'. Cannot display content as text."
except requests.exceptions.RequestException as e: return f"Error downloading or reading file: {e}"
@tool
def python_interpreter(code: str) -> str:
"""Executes Python code and returns its stdout."""
print(f"--- Calling Python Interpreter Tool with code:\n{code} ---")
output_buffer = io.StringIO()
try:
with redirect_stdout(output_buffer): exec(code, globals())
return f"Code executed successfully. Output:\n{output_buffer.getvalue()}"
except Exception as e: return f"Error executing Python code: {e}"
##================================================================================================
#✅ 2. CONFIGURE AND BUILD THE AGENT (with Qwen2 and Manual Tool Calling)
#================================================================================================
class AgentState(TypedDict):
messages: Annotated[List[BaseMessage], operator.add]
def build_agent_graph():
"""Builds the agent using a manual LangGraph loop with the HuggingFaceEndpoint."""
tools = [tavily_search, read_file, python_interpreter]
tool_map = {tool.name: tool for tool in tools}
Generated code
# Using Qwen2-72B-Instruct model via HuggingFaceEndpoint
repo_id = "Qwen/Qwen2-72B-Instruct"
llm = HuggingFaceEndpoint(
    repo_id=repo_id,
    max_new_tokens=1024,
    temperature=0.1,
    huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
)

def call_model(state: AgentState):
    """Invokes the LLM and wraps the response in an AIMessage."""
    # Qwen2 Instruct uses a specific chat template. We build it manually.
    prompt_str = ""
    for msg in state['messages']:
        role = ""
        if isinstance(msg, SystemMessage): role = "system"
        elif isinstance(msg, HumanMessage): role = "user"
        elif isinstance(msg, AIMessage): role = "assistant"
        elif isinstance(msg, ToolMessage): continue # We'll handle tool results differently

        if role: prompt_str += f"<|im_start|>{role}\n{msg.content}<|im_end|>\n"

    # Add results from the last tool call, if any
    if isinstance(state['messages'][-1], ToolMessage):
         prompt_str += f"<|im_start|>user\nTool output:\n{state['messages'][-1].content}<|im_end|>\n"

    prompt_str += "<|im_start|>assistant\n"

    response_text = llm.invoke(prompt_str)
    return {"messages": [AIMessage(content=response_text)]}

def should_continue(state: AgentState) -> str:
    """Determines whether to call a tool or end the loop."""
    last_message_content = state['messages'][-1].content.strip()
    # A simple check for JSON is a reliable way to detect tool calls.
    if "```json" in last_message_content:
        return "action"
    if last_message_content.startswith('{') and last_message_content.endswith('}'):
        try:
            json.loads(last_message_content)
            return "action"
        except json.JSONDecodeError:
            return "end" # Not valid JSON, must be the final answer
    else:
        return "end"

def call_tool_node(state: AgentState):
    """Parses the JSON tool call from the LLM and executes it."""
    last_message_content = state['messages'][-1].content.strip()
    
    # Extract JSON from markdown code block if present
    if "```json" in last_message_content:
        json_str = last_message_content.split("```json").split("```")[0].strip()
    else:
        json_str = last_message_content

    try:
        tool_call_data = json.loads(json_str)
        tool_name = tool_call_data.get("tool_name")
        parameters = tool_call_data.get("parameters", {})
        if tool_name not in tool_map:
            return {"messages": [ToolMessage(content=f"Error: Tool '{tool_name}' not found.", tool_call_id="error")]}

        selected_tool = tool_map[tool_name]
        tool_output = selected_tool.invoke(parameters)
        return {"messages": [ToolMessage(content=str(tool_output), tool_call_id=tool_name)]}
    except Exception as e:
        return {"messages": [ToolMessage(content=f"Error parsing tool call: {e}. Content: '{last_message_content}'", tool_call_id="error")]}

workflow = StateGraph(AgentState)
workflow.add_node("agent", call_model)
workflow.add_node("action", call_tool_node)
workflow.set_entry_point("agent")
workflow.add_conditional_edges("agent", should_continue, {"action": "action", "end": END})
workflow.add_edge('action', 'agent')
return workflow.compile()
Use code with caution.
#================================================================================================
#✅ 3. AGENT CLASS AND EVALUATION LOGIC
#================================================================================================
class GaiaAgent:
def init(self):
print("GaiaAgent initialized. Building agent with Qwen/Qwen2-72B-Instruct...")
self.agent_app = build_agent_graph()
Generated code
def __call__(self, question: str) -> str:
    print(f"\n{'='*60}\nAgent received question: {question[:100]}...\n{'='*60}")
    try:
        initial_input = {"messages": [SystemMessage(content=AGENT_SYSTEM_PROMPT), HumanMessage(content=question)]}
        final_state = None
        for step in self.agent_app.stream(initial_input, {"recursion_limit": 15}):
            final_state = list(step.values())[0]

        final_answer = final_state['messages'][-1].content
        return str(final_answer).strip()
    except Exception as e:
        print(f"An error occurred during agent execution: {e}")
        return f"AGENT_EXECUTION_ERROR: {e}"
Use code with caution.
--- The rest of the file is unchanged ---
def run_and_submit_all( profile: gr.OAuthProfile | None):
space_id = os.getenv("SPACE_ID")
if not profile: return "Please Login to Hugging Face with the button.", None
username = f"{profile.username}"
print(f"User logged in: {username}")
api_url = DEFAULT_API_URL
questions_url = f"{api_url}/questions"
submit_url = f"{api_url}/submit"
agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
Generated code
try:
    response = requests.get(questions_url, timeout=15)
    response.raise_for_status()
    questions_data = response.json()
except Exception as e: return f"An unexpected error occurred fetching questions: {e}", None

results_log, answers_payload = [], []
agent_instance = GaiaAgent()

for item in questions_data:
    task_id, question_text = item.get("task_id"), item.get("question")
    if not task_id or question_text is None: continue
    try:
        submitted_answer = agent_instance(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: return "Agent did not produce any answers to submit.", pd.DataFrame(results_log)
submission_data = {"username": username.strip(), "agent_code": agent_code, "answers": answers_payload}

try:
    response = requests.post(submit_url, json=submission_data, timeout=90)
    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.')}"
    )
    return final_status, pd.DataFrame(results_log)
except Exception as e: return f"An unexpected error in submission: {e}", pd.DataFrame(results_log)
Use code with caution.
with gr.Blocks() as demo:
gr.Markdown("# GAIA Agent Final Assessment (Qwen2-72B-Instruct)")
gr.Markdown(
"""
Instructor's Note: This version uses the powerful Qwen/Qwen2-72B-Instruct model from the Hugging Face Hub.
It relies on a robust manual LangGraph loop to handle tool calls by instructing the model to generate JSON.
1. Ensure you have a HUGGINGFACEHUB_API_TOKEN and TAVILY_API_KEY set in your secrets.
2. Ensure your requirements.txt is updated. Good luck!
"""
)
gr.LoginButton()
run_button = gr.Button("Run Evaluation & Submit All Answers")
status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
if name == "main":
print("\n" + "-"*30 + " App Starting " + "-"*30)
demo.launch(debug=True, share=False, ssr_mode=False)