import os import gradio as gr import requests import inspect import pandas as pd from langgraph.prebuilt import ToolNode, tools_condition from langgraph.graph.message import add_messages from typing import List, Dict from typing import TypedDict, Annotated from langchain_core.messages import AnyMessage, HumanMessage from langgraph.graph import START, StateGraph from langchain_google_genai import ChatGoogleGenerativeAI from ToolSet import toolset from Utils.final_answer import extract_final_answer from Utils.handle_file import handle_attachment from fetch_question import get_all_questions, get_one_random_question, submit # (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 ------ gemini_api_key = os.getenv("GEMINI_API_KEY") tivaly_api_key = os.getenv("TAVILY_API_KEY") llm = ChatGoogleGenerativeAI( model="gemini-2.0-flash", temperature=0, google_api_key = gemini_api_key ) llm_with_tools = llm.bind_tools(toolset) sys_prompt_file = open("sys_prompt.txt") SYSTEM_PROMPT = sys_prompt_file.read() class AgentState(TypedDict): messages: Annotated[list[AnyMessage], add_messages] def assistant(state: AgentState): return { "messages": [llm_with_tools.invoke(state["messages"])], } builder = StateGraph(AgentState) builder.add_node("assistant", assistant) builder.add_node("tools", ToolNode(toolset)) builder.add_edge(START, "assistant") builder.add_conditional_edges( "assistant", tools_condition ) builder.add_edge("tools","assistant") gaia_agent = builder.compile() def run_and_submit_all(profile: gr.OAuthProfile | None): """ Fetches all questions, runs the agent on them, submits all answers, and displays the results. Handles attachments if present. """ # --- 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 # 1. Instantiate Agent (modify this part to create your agent) try: agent = gaia_agent 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 (useful for others so please keep it public) agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main" print(agent_code) # 2. Fetch Questions questions_data = get_all_questions() # 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 # 2.2 Handle attachment if present attachment_info = None if "file_name" in item and item["file_name"]: file_name = item.get("file_name") attachment_info = handle_attachment(task_id, file_name) print(f"Attachment handling result: {attachment_info['status']}") try: # Prepare messages based on attachment handling messages = [ SystemMessage(content=SYSTEM_PROMPT), SystemMessage(content=f"Current task id: {task_id}") ] # If we have an attachment that Claude can process directly if attachment_info and attachment_info["status"] == "success" and attachment_info["handling"] == "direct": # Encode content for direct inclusion encoded_content = base64.b64encode(attachment_info["raw_content"]).decode('utf-8') content_type = attachment_info["content_type"] # Create multimodal message if content_type.startswith('image/'): multimodal_content = [ {"type": "text", "text": question_text}, { "type": "image", "source": { "type": "base64", "media_type": content_type, "data": encoded_content } } ] elif content_type == "application/pdf" or "spreadsheet" in content_type or "excel" in content_type or "csv" in content_type: multimodal_content = [ {"type": "text", "text": question_text}, { "type": "file", "source": { "type": "base64", "media_type": content_type, "data": encoded_content }, "name": attachment_info["file_name"] } ] messages.append(HumanMessage(content=multimodal_content)) # If we have an attachment that needs tool processing elif attachment_info and attachment_info["status"] == "success" and attachment_info["handling"] == "tool": # Add info about the file to the question file_info = ( f"{question_text}\n\n" f"Note: This task has an attached file that can be accessed at: {attachment_info['file_path']}\n" f"File type: {attachment_info['content_type']}" ) messages.append(HumanMessage(content=file_info)) # If no attachment or error with attachment else: messages.append(HumanMessage(content=question_text)) # Invoke the agent with the prepared messages agent_answer = agent.invoke({"messages": messages},{"recursion_limit": 50}) submitted_answer = extract_final_answer(agent_answer['messages'][-1].content) 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 return submit(submission_data, results_log) def run_and_submit_one( profile: gr.OAuthProfile | None): # --- 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 # 1. Instantiate Agent ( modify this part to create your agent) try: agent = gaia_agent 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 questions_data = get_one_random_question() print("questions_data:", questions_data) # 2.2 Handle attachment if present attachment_info = None if "file_name" in questions_data and questions_data["file_name"]: task_id = questions_data.get("task_id") file_name = questions_data.get("file_name") attachment_info = handle_attachment(task_id, file_name) print(f"Attachment handling result: {attachment_info['status']}") # 3. Run your Agent results_log = [] answers_payload = [] print(f"Running agent on {len(questions_data)} questions...") task_id = questions_data.get("task_id") question_text = questions_data.get("question") if not task_id or question_text is None: print(f"Skipping item with missing task_id or question") try: # Prepare messages based on attachment handling messages = [ SystemMessage(content=SYSTEM_PROMPT), SystemMessage(content=f"Current task id: {task_id}") ] # If we have an attachment that Claude can process directly if attachment_info and attachment_info["status"] == "success" and attachment_info["handling"] == "direct": # Encode content for direct inclusion encoded_content = base64.b64encode(attachment_info["raw_content"]).decode('utf-8') content_type = attachment_info["content_type"] # Create multimodal message if content_type.startswith('image/'): multimodal_content = [ {"type": "text", "text": question_text}, { "type": "image", "source": { "type": "base64", "media_type": content_type, "data": encoded_content } } ] elif content_type == "application/pdf" or "spreadsheet" in content_type or "excel" in content_type or "csv" in content_type: multimodal_content = [ {"type": "text", "text": question_text}, { "type": "file", "source": { "type": "base64", "media_type": content_type, "data": encoded_content }, "name": attachment_info["file_name"] } ] messages.append(HumanMessage(content=multimodal_content)) # If we have an attachment that needs tool processing elif attachment_info and attachment_info["status"] == "success" and attachment_info["handling"] == "tool": # Add info about the file to the question file_info = ( f"{question_text}\n\n" f"Note: This task has an attached file that can be accessed at: {attachment_info['file_path']}\n" f"File type: {attachment_info['content_type']}" ) messages.append(HumanMessage(content=file_info)) # If no attachment or error with attachment else: messages.append(HumanMessage(content=question_text)) # Invoke the agent with the prepared messages agent_answer = agent.invoke({"messages": messages},{"recursion_limit": 50}) submitted_answer = extract_final_answer(agent_answer['messages'][-1].content) 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) return submit(submission_data, results_log) # --- 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") run_one_button = gr.Button("Run one question and submit") 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] ) run_one_button.click( fn=run_and_submit_one, 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)