Update app.py

app.py

@@ -1,34 +1,246 @@
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
+import re
+import json
+import math
+import unicodedata
+from datetime import datetime
+
+# --- LangGraph + LangChain imports ---
+from langgraph.prebuilt import create_react_agent
+from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
+from langchain_core.tools import tool
+from langchain_community.tools import DuckDuckGoSearchRun
+from langchain_community.utilities import WikipediaAPIWrapper
+from langchain_core.messages import SystemMessage
 
-# (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 ------
+# ─────────────────────────────────────────────
+#  TOOLS
+# ─────────────────────────────────────────────
+
+@tool
+def web_search(query: str) -> str:
+    """Search the web using DuckDuckGo. Use for current events, facts, and general knowledge."""
+    try:
+        search = DuckDuckGoSearchRun()
+        return search.run(query)
+    except Exception as e:
+        return f"Search error: {e}"
+
+
+@tool
+def wikipedia_search(query: str) -> str:
+    """Search Wikipedia for encyclopedic knowledge, historical facts, biographies, science."""
+    try:
+        wiki = WikipediaAPIWrapper(top_k_results=2, doc_content_chars_max=3000)
+        return wiki.run(query)
+    except Exception as e:
+        return f"Wikipedia error: {e}"
+
+
+@tool
+def python_repl(code: str) -> str:
+    """
+    Execute Python code for math calculations, data processing, logic.
+    Always print() the final result.
+    Example: print(2 + 2)
+    """
+    import io, sys, math, json, re, unicodedata, datetime
+    old_stdout = sys.stdout
+    sys.stdout = io.StringIO()
+    try:
+        exec(code, {
+            "math": math, "json": json, "re": re,
+            "unicodedata": unicodedata, "datetime": datetime,
+            "__builtins__": __builtins__
+        })
+        output = sys.stdout.getvalue()
+        return output.strip() if output.strip() else "Code executed (no output). Use print() to see results."
+    except Exception as e:
+        return f"Code error: {e}"
+    finally:
+        sys.stdout = old_stdout
+
+
+@tool
+def read_file_from_url(url: str) -> str:
+    """
+    Download and read a file from a URL (txt, csv, json, py, etc.).
+    Returns the file content as text.
+    """
+    try:
+        response = requests.get(url, timeout=15)
+        response.raise_for_status()
+        content_type = response.headers.get("Content-Type", "")
+        if "text" in content_type or "json" in content_type:
+            return response.text[:5000]
+        else:
+            return f"Binary file ({content_type}), cannot read as text."
+    except Exception as e:
+        return f"Error reading file: {e}"
+
+
+@tool
+def get_task_file(task_id: str) -> str:
+    """
+    Fetch the file associated with a GAIA task by its task_id.
+    Returns file content or description.
+    """
+    try:
+        api_url = "https://agents-course-unit4-scoring.hf.space"
+        url = f"{api_url}/files/{task_id}"
+        response = requests.get(url, timeout=15)
+        if response.status_code == 200:
+            content_type = response.headers.get("Content-Type", "")
+            if "text" in content_type or "json" in content_type:
+                return response.text[:5000]
+            elif "image" in content_type:
+                return f"[Image file attached to task {task_id} - content-type: {content_type}]"
+            elif "audio" in content_type:
+                return f"[Audio file attached to task {task_id} - content-type: {content_type}]"
+            else:
+                return f"[File attached: {content_type}]"
+        else:
+            return f"No file found for task {task_id}"
+    except Exception as e:
+        return f"Error fetching task file: {e}"
+
+
+@tool
+def calculator(expression: str) -> str:
+    """
+    Evaluate a simple math expression safely.
+    Examples: '2 + 2', '100 * 1.07 ** 5', 'math.sqrt(144)'
+    """
+    try:
+        result = eval(expression, {"math": math, "__builtins__": {}})
+        return str(result)
+    except Exception as e:
+        return f"Calculation error: {e}. Try python_repl for complex code."
+
+
+# ─────────────────────────────────────────────
+#  SYSTEM PROMPT
+# ─────────────────────────────────────────────
+
+SYSTEM_PROMPT = """You are a precise, expert AI assistant solving GAIA benchmark questions.
+
+GAIA questions require careful reasoning and often multiple steps. Follow these rules:
+
+## Answer Format (CRITICAL)
+- Your FINAL answer must be the **bare minimum**: a number, a word, a name, a date, a short phrase.
+- NO explanations, NO punctuation at the end, NO "The answer is...", NO sentences.
+- Examples of correct final answers: `42`, `Marie Curie`, `Paris`, `1969`, `blue`, `$14.50`
+- For lists, separate items with commas: `item1, item2, item3`
+
+## Strategy
+1. **Read carefully** – identify exactly what is being asked.
+2. **Use tools** – search the web, Wikipedia, or run code to verify facts.
+3. **Verify numbers** – always double-check calculations with the calculator or python_repl.
+4. **Check for files** – if the question mentions an attachment or file, use get_task_file.
+5. **Be specific** – GAIA answers are exact; approximate answers are wrong.
+
+## Tool Usage
+- Use `web_search` for recent events, facts, and general knowledge.
+- Use `wikipedia_search` for biographies, history, science.
+- Use `python_repl` for calculations, data manipulation, logic puzzles.
+- Use `calculator` for quick arithmetic.
+- Use `get_task_file` when a question refers to an attached file or document.
+
+## Final Answer
+Always end your response with:
+FINAL ANSWER: <your answer here>
+"""
+
+# ─────────────────────────────────────────────
+#  AGENT
+# ─────────────────────────────────────────────
+
 class BasicAgent:
     def __init__(self):
-        print("BasicAgent initialized.")
+        print("Initializing LangGraph ReAct Agent with Llama 3.3 70B...")
+
+        hf_token = os.getenv("HF_TOKEN")
+
+        llm_endpoint = HuggingFaceEndpoint(
+            repo_id="meta-llama/Llama-3.3-70B-Instruct",
+            huggingfacehub_api_token=hf_token,
+            task="text-generation",
+            max_new_tokens=1024,
+            temperature=0.1,
+            do_sample=False,
+        )
+        llm = ChatHuggingFace(llm=llm_endpoint)
+
+        tools = [
+            web_search,
+            wikipedia_search,
+            python_repl,
+            calculator,
+            read_file_from_url,
+            get_task_file,
+        ]
+
+        self.agent = create_react_agent(
+            model=llm,
+            tools=tools,
+            state_modifier=SYSTEM_PROMPT,
+        )
+
+        print("Agent ready.")
+
     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
+        print(f"\n[AGENT] Question: {question[:100]}...")
+        try:
+            result = self.agent.invoke({
+                "messages": [("user", question)]
+            })
 
-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
+            # Extract last AI message
+            last_message = result["messages"][-1].content
+            print(f"[AGENT] Raw output: {last_message[:200]}...")
+
+            # Extract FINAL ANSWER if present
+            answer = self._extract_final_answer(last_message)
+            print(f"[AGENT] Final answer: {answer}")
+            return answer
+
+        except Exception as e:
+            print(f"[AGENT] Error: {e}")
+            return f"Error: {e}"
+
+    def _extract_final_answer(self, text: str) -> str:
+        """Extract the FINAL ANSWER from agent output."""
+        # Try to find "FINAL ANSWER: ..." pattern
+        patterns = [
+            r"FINAL ANSWER:\s*(.+?)(?:\n|$)",
+            r"Final Answer:\s*(.+?)(?:\n|$)",
+            r"final answer:\s*(.+?)(?:\n|$)",
+        ]
+        for pattern in patterns:
+            match = re.search(pattern, text, re.IGNORECASE)
+            if match:
+                return match.group(1).strip()
+
+        # Fallback: return last non-empty line
+        lines = [l.strip() for l in text.strip().split("\n") if l.strip()]
+        return lines[-1] if lines else text.strip()
+
+
+# ─────────────────────────────────────────────
+#  GRADIO RUNNER
+# ─────────────────────────────────────────────
+
+def run_and_submit_all(profile: gr.OAuthProfile | None):
+    space_id = os.getenv("SPACE_ID")
 
     if profile:
-        username= f"{profile.username}"
+        username = f"{profile.username}"
         print(f"User logged in: {username}")
     else:
         print("User not logged in.")
@@ -38,15 +250,15 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
 
-    # 1. Instantiate Agent ( modify this part to create your agent)
+    # 1. Init 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)
+    print(f"Agent code: {agent_code}")
 
     # 2. Fetch Questions
     print(f"Fetching questions from: {questions_url}")
@@ -55,24 +267,16 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         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
+            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
+        return f"Error fetching questions: {e}", None
 
-    # 3. Run your Agent
+    # 3. Run 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")
@@ -82,24 +286,32 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         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})
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text[:100],
+                "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}"})
+            print(f"Error on task {task_id}: {e}")
+            results_log.append({
+                "Task ID": task_id,
+                "Question": question_text[:100],
+                "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)
+    # 4. Submit
+    submission_data = {
+        "username": username.strip(),
+        "agent_code": agent_code,
+        "answers": answers_payload
+    }
+    print(f"Submitting {len(answers_payload)} answers...")
 
-    # 5. Submit
-    print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     try:
-        response = requests.post(submit_url, json=submission_data, timeout=60)
+        response = requests.post(submit_url, json=submission_data, timeout=120)
         response.raise_for_status()
         result_data = response.json()
         final_status = (
@@ -110,60 +322,41 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
             f"Message: {result_data.get('message', 'No message received.')}"
         )
         print("Submission successful.")
-        results_df = pd.DataFrame(results_log)
-        return final_status, results_df
+        return final_status, pd.DataFrame(results_log)
     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:
+        except Exception:
             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
+        return f"Submission Failed: {error_detail}", pd.DataFrame(results_log)
     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
+        return f"Submission Failed: {e}", pd.DataFrame(results_log)
 
 
-# --- Build Gradio Interface using Blocks ---
+# ─────────────────────────────────────────────
+#  GRADIO UI
+# ─────────────────────────────────────────────
+
 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.Markdown("# 🤖 GAIA Agent — LangGraph + Llama 3.3 70B")
+    gr.Markdown("""
+**Stack:** LangGraph ReAct · Llama 3.3 70B (HF Inference) · DuckDuckGo · Wikipedia · Python REPL
+
+**Instructions:**
+1. Log in with your HuggingFace account below.
+2. Make sure `HF_TOKEN` is set as a Space secret (with access to Llama 3.3 70B).
+3. Click **Run Evaluation & Submit All Answers**.
+
+> ⚠️ The run can take several minutes — the agent reasons through each question step by step.
+    """)
 
     gr.LoginButton()
 
-    run_button = gr.Button("Run Evaluation & Submit All Answers")
+    run_button = gr.Button("▶️ Run Evaluation & Submit All Answers", variant="primary")
 
-    status_output = gr.Textbox(label="Run Status / Submission Result", lines=5, interactive=False)
-    # Removed max_rows=10 from DataFrame constructor
+    status_output = gr.Textbox(label="Run Status / Submission Result", lines=6, interactive=False)
     results_table = gr.DataFrame(label="Questions and Agent Answers", wrap=True)
 
     run_button.click(
@@ -172,25 +365,13 @@ with gr.Blocks() as demo:
     )
 
 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...")
+    print("\n" + "-" * 30 + " App Starting " + "-" * 30)
+    space_host = os.getenv("SPACE_HOST")
+    space_id = os.getenv("SPACE_ID")
+    if space_host:
+        print(f"✅ SPACE_HOST: {space_host}")
+    if space_id:
+        print(f"✅ SPACE_ID: {space_id}")
+        print(f"   Repo: https://huggingface.co/spaces/{space_id}/tree/main")
+    print("-" * 60 + "\n")
     demo.launch(debug=True, share=False)