Upload 6 files

agent.py

@@ -0,0 +1,144 @@
+# agent.py
+import os
+from dotenv import load_dotenv
+from langchain_openai import ChatOpenAI
+from langchain_core.messages import SystemMessage, HumanMessage
+from langgraph.graph import StateGraph, START, END
+from langgraph.prebuilt import ToolNode, tools_condition
+
+from tools import TOOLS
+from typing import TypedDict, Annotated, List
+from langchain_core.messages import BaseMessage
+import operator
+
+load_dotenv()
+
+# ── 1. State Definition ─────────────────────────────────────────────────────
+
+class AgentState(TypedDict):
+    """
+    This is the shared state that flows through the entire graph.
+    Every node can read from it and write to it.
+    
+    - messages: The conversation history (the LLM sees all of this).
+                operator.add means new messages are APPENDED, not replaced.
+    - task_id:  The GAIA question ID (needed to download attached files).
+    """
+    messages: Annotated[List[BaseMessage], operator.add]
+    task_id: str
+
+# ── 2. System Prompt ────────────────────────────────────────────────────────
+
+SYSTEM_PROMPT = """You are a precise research assistant solving benchmark questions.
+
+RULES:
+- You MUST use the available tools to find accurate information. Do NOT guess.
+- After finding the answer, respond with ONLY the answer itself.
+- Do NOT include phrases like "The answer is", "Based on my research", or "FINAL ANSWER".
+- Do NOT include units unless they are explicitly part of the answer.
+- For numbers, give the exact number (e.g., "42" not "approximately 42").
+- For names, give the full correct name with correct spelling.
+- For dates, match the format implied by the question (e.g., "1969" or "July 20, 1969").
+
+Available tools: web_search, wikipedia_search, calculate, read_excel_file.
+Choose the right tool for the question type.
+"""
+
+# ── 3. LLM with Tools Bound ─────────────────────────────────────────────────
+
+llm = ChatOpenAI(
+    model="gpt-4o-mini",          # Affordable and capable; swap to gpt-4o for better results
+    temperature=0,                 # Zero temperature = deterministic, no creative hallucinations
+    api_key=os.getenv("OPENAI_API_KEY")
+)
+
+# Binding tools to the LLM tells it what tools are available and how to call them
+llm_with_tools = llm.bind_tools(TOOLS)
+
+# ── 4. Node Definitions ─────────────────────────────────────────────────────
+
+def agent_node(state: AgentState) -> dict:
+    """
+    The main thinking node. The LLM looks at all messages so far
+    and decides: call a tool, or give the final answer.
+    """
+    # Prepend the system message to give the LLM its instructions
+    messages = [SystemMessage(content=SYSTEM_PROMPT)] + state["messages"]
+    
+    # Call the LLM (which has tools bound to it)
+    response = llm_with_tools.invoke(messages)
+    
+    # Return the response as a new message to be appended to state
+    return {"messages": [response]}
+
+
+# ToolNode is a pre-built LangGraph node that automatically handles tool calls.
+# It reads the tool_calls from the last AI message, runs the tools, and returns results.
+tool_node = ToolNode(tools=TOOLS)
+
+# ── 5. Routing Logic ────────────────────────────────────────────────────────
+
+# tools_condition is a pre-built LangGraph function that checks if the last
+# AI message contains tool calls. If yes → go to "tools"; if no → go to END.
+# This is the conditional edge that creates the ReAct loop.
+
+# ── 6. Build the Graph ──────────────────────────────────────────────────────
+
+def build_agent():
+    """Constructs and compiles the LangGraph agent."""
+    
+    graph = StateGraph(AgentState)
+    
+    # Add nodes
+    graph.add_node("agent", agent_node)
+    graph.add_node("tools", tool_node)
+    
+    # Entry point: always start at the agent node
+    graph.add_edge(START, "agent")
+    
+    # Conditional edge: after the agent node, check if a tool was called
+    graph.add_conditional_edges(
+        "agent",          # Source node
+        tools_condition,  # The function that decides where to go
+        # tools_condition returns "tools" if tool_calls exist, else "__end__"
+    )
+    
+    # After tools run, always go back to the agent (to process tool results)
+    graph.add_edge("tools", "agent")
+    
+    # Compile the graph into a runnable
+    return graph.compile()
+
+
+# Create the agent (call this once at startup)
+agent = build_agent()
+
+
+# ── 7. Run Function ─────────────────────────────────────────────────────────
+
+def run_agent(question: str, task_id: str = "") -> str:
+    """
+    Runs the agent on a single GAIA question.
+    
+    Args:
+        question: The question text from the GAIA API.
+        task_id:  The task ID (needed if a file is attached).
+    
+    Returns:
+        The agent's answer as a string.
+    """
+    initial_state = {
+        "messages": [HumanMessage(content=question)],
+        "task_id": task_id
+    }
+    
+    # recursion_limit prevents infinite loops (default is 25)
+    config = {"recursion_limit": 15}
+    
+    try:
+        final_state = agent.invoke(initial_state, config=config)
+        # The last message in the state is the agent's final answer
+        last_message = final_state["messages"][-1]
+        return last_message.content.strip()
+    except Exception as e:
+        return f"Agent error: {str(e)}"

api_client.py

@@ -0,0 +1,42 @@
+# api_client.py
+import requests
+
+SCORING_API = "https://agents-course-unit4-scoring.hf.space"
+
+def get_all_questions() -> list[dict]:
+    """
+    Fetches all 20 Level 1 questions from the GAIA scoring API.
+    
+    Returns a list of dicts, each with:
+        - task_id: str  (unique ID for the question)
+        - question: str (the question text)
+        - file_name: str (filename of attached file, if any; else empty string)
+    """
+    response = requests.get(f"{SCORING_API}/questions", timeout=30)
+    if response.status_code == 200:
+        return response.json()
+    else:
+        raise Exception(f"Failed to fetch questions: {response.status_code}")
+
+def submit_answers(username: str, agent_code_url: str, answers: list[dict]) -> dict:
+    """
+    Submits answers to the GAIA API for scoring.
+    
+    Args:
+        username:       Your HuggingFace username.
+        agent_code_url: URL to your HF Space code (e.g. https://huggingface.co/spaces/yourname/space/tree/main)
+        answers:        List of {"task_id": "...", "submitted_answer": "..."} dicts.
+    
+    Returns:
+        A dict with your score and leaderboard position.
+    """
+    payload = {
+        "username": username,
+        "agent_code": agent_code_url,
+        "answers": answers
+    }
+    response = requests.post(f"{SCORING_API}/submit", json=payload, timeout=60)
+    if response.status_code == 200:
+        return response.json()
+    else:
+        raise Exception(f"Submission failed: {response.status_code} — {response.text}")

app.py

@@ -1,196 +1,91 @@
-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)
+# app.py
+import gradio as gr
+import pandas as pd
+from agent import run_agent
+from api_client import get_all_questions, submit_answers
+
+def run_evaluation(profile: gr.OAuthProfile | None):
+    """
+    The main function triggered when the user clicks "Run Evaluation".
+    Fetches all questions, runs the agent on each, and submits answers.
+    """
+    if profile is None:
+        return "Please log in with your HuggingFace account first.", None
+    
+    username = profile.username
+    agent_code_url = f"https://huggingface.co/spaces/{username}/gaia-agent/tree/main"
+    
+    # Step 1: Fetch questions
+    try:
+        questions = get_all_questions()
+    except Exception as e:
+        return f"Failed to fetch questions: {e}", None
+    
+    # Step 2: Run the agent on each question
+    answers = []
+    results_log = []  # For displaying in the UI table
+    
+    for q in questions:
+        task_id = q["task_id"]
+        question_text = q["question"]
+        
+        print(f"\n{'='*60}")
+        print(f"Question: {question_text[:100]}...")
+        
+        # Run the agent
+        answer = run_agent(question=question_text, task_id=task_id)
+        
+        print(f"Answer: {answer}")
+        
+        answers.append({
+            "task_id": task_id,
+            "submitted_answer": answer
+        })
+        
+        results_log.append({
+            "Task ID": task_id[:8] + "...",
+            "Question (truncated)": question_text[:80] + "...",
+            "Agent Answer": answer
+        })
+    
+    # Step 3: Submit answers
+    try:
+        result = submit_answers(username, agent_code_url, answers)
+        score = result.get("score", "N/A")
+        correct = result.get("correct_count", "N/A")
+        total = result.get("total_attempted", len(answers))
+        
+        summary = f"✅ Submitted! Score: {score:.1%} ({correct}/{total} correct)"
+    except Exception as e:
+        summary = f"⚠️ Submission error: {e}"
+    
+    # Return summary text and a dataframe for display
+    df = pd.DataFrame(results_log)
+    return summary, df
+
+
+# ── Build the Gradio Interface ───────────────────────────────────────────────
+
+with gr.Blocks() as demo:
+    gr.Markdown("# GAIA Level 1 Agent — HuggingFace Agents Course")
+    gr.Markdown(
+        "Log in with your HuggingFace account, then click **Run Evaluation** "
+        "to run the agent on all 20 GAIA Level 1 questions and submit your score."
+    )
+    
+    # HuggingFace OAuth login button
+    login_btn = gr.LoginButton()
+    
+    run_btn = gr.Button("Run Evaluation", variant="primary")
+    
+    status_output = gr.Textbox(label="Status", interactive=False)
+    results_table = gr.Dataframe(label="Results", interactive=False)
+    
+    run_btn.click(
+        fn=run_evaluation,
+        inputs=[],  # profile is injected automatically by Gradio
+        outputs=[status_output, results_table]
+    )
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -1,2 +1,11 @@
-gradio
-requests
+langgraph>=0.2.0
+langchain>=0.3.0
+langchain-openai>=0.2.0
+langchain-community>=0.3.0
+tavily-python>=0.3.0
+gradio>=5.0.0
+requests>=2.31.0
+pandas>=2.0.0
+openpyxl>=3.1.0
+duckduckgo-search>=6.0.0
+python-dotenv>=1.0.0

tools.py

@@ -0,0 +1,117 @@
+# tools.py
+import os
+import requests
+import pandas as pd
+from langchain_core.tools import tool
+from langchain_community.tools import DuckDuckGoSearchRun
+
+# ── Tool 1: Web Search ──────────────────────────────────────────────────────
+
+@tool
+def web_search(query: str) -> str:
+    """
+    Search the web for current information.
+    Use this for facts, recent events, or anything not in your training data.
+    
+    Args:
+        query: The search query string.
+    Returns:
+        A string containing the top search results.
+    """
+    # Try Tavily first (better quality), fall back to DuckDuckGo
+    tavily_key = os.getenv("TAVILY_API_KEY")
+    if tavily_key:
+        from tavily import TavilyClient
+        client = TavilyClient(api_key=tavily_key)
+        results = client.search(query=query, max_results=3)
+        # Format results into a readable string
+        return "\n\n".join([
+            f"Source: {r['url']}\n{r['content']}"
+            for r in results.get("results", [])
+        ])
+    else:
+        # DuckDuckGo fallback
+        search = DuckDuckGoSearchRun()
+        return search.run(query)
+
+
+# ── Tool 2: Wikipedia Lookup ────────────────────────────────────────────────
+
+@tool
+def wikipedia_search(query: str) -> str:
+    """
+    Look up factual information on Wikipedia.
+    Prefer this over web_search for well-established facts, historical events,
+    or definitions of concepts.
+    
+    Args:
+        query: The topic or person to search for.
+    Returns:
+        A summary from Wikipedia.
+    """
+    from langchain_community.utilities import WikipediaAPIWrapper
+    wiki = WikipediaAPIWrapper(top_k_results=2, doc_content_chars_max=2000)
+    return wiki.run(query)
+
+
+# ── Tool 3: Calculator / Python Evaluator ───────────────────────────────────
+
+@tool
+def calculate(expression: str) -> str:
+    """
+    Evaluate a mathematical expression.
+    Use this for any arithmetic, unit conversions, or numerical reasoning.
+    
+    Args:
+        expression: A valid Python math expression, e.g. "2 ** 10 / 1024"
+    Returns:
+        The result as a string.
+    """
+    try:
+        # Safe evaluation: only allow math operations
+        import math
+        allowed = {k: getattr(math, k) for k in dir(math) if not k.startswith('_')}
+        result = eval(expression, {"__builtins__": {}}, allowed)
+        return str(result)
+    except Exception as e:
+        return f"Error evaluating expression: {e}"
+
+
+# ── Tool 4: Read Spreadsheet Files ──────────────────────────────────────────
+
+@tool
+def read_excel_file(task_id: str) -> str:
+    """
+    Download and read a spreadsheet (.xlsx or .csv) attached to a GAIA question.
+    Use this when the question references a file or asks about data in a table.
+    
+    Args:
+        task_id: The GAIA task ID for the question (not the filename).
+    Returns:
+        A string representation of the file's contents.
+    """
+    api_url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}"
+    response = requests.get(api_url)
+    
+    if response.status_code != 200:
+        return f"Could not download file for task {task_id}."
+    
+    # Detect file type from headers or content
+    content_type = response.headers.get("content-type", "")
+    
+    try:
+        import io
+        if "csv" in content_type or task_id.endswith(".csv"):
+            df = pd.read_csv(io.BytesIO(response.content))
+        else:
+            df = pd.read_excel(io.BytesIO(response.content))
+        
+        # Return a text representation (first 50 rows)
+        return f"File contents (first 50 rows):\n{df.head(50).to_string()}"
+    except Exception as e:
+        return f"Error reading file: {e}"
+
+
+# ── Collect all tools into a list ───────────────────────────────────────────
+
+TOOLS = [web_search, wikipedia_search, calculate, read_excel_file]