Update app.py

app.py

@@ -1,73 +1,95 @@
 import os
 import gradio as gr
 import requests
-import inspect
 import pandas as pd
 from groq import Groq
 
+# --- New Imports for LangChain Agent ---
+from langchain_groq import ChatGroq
+from langchain.agents import AgentExecutor, create_tool_calling_agent
+from langchain_community.tools.tavily_search import TavilySearchResults
+from langchain_core.prompts import ChatPromptTemplate
+
+
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
-# --- Groq Powered Agent Definition ---
-# This new agent uses the Groq API to generate answers.
-class GroqAgent:
-    def __init__(self, api_key):
+
+# --- Agent Definition ---
+# This new agent uses LangChain to orchestrate an LLM with tools.
+class LangChainAgent:
+    def __init__(self, groq_api_key, tavily_api_key):
         """
-        Initializes the GroqAgent with the Groq API client.
+        Initializes the agent with an LLM and a set of tools.
         """
-        print("Initializing GroqAgent...")
-        self.client = Groq(api_key=api_key)
-        print("GroqAgent initialized successfully.")
+        print("Initializing LangChainAgent...")
+
+        # 1. Initialize the LLM
+        # We use ChatGroq, the LangChain integration for Groq's API.
+        self.llm = ChatGroq(
+            model_name="llama3-70b-8192",
+            groq_api_key=groq_api_key,
+            temperature=0.0
+        )
+
+        # 2. Define the tools the agent can use
+        # For now, we'll just give it a web search tool.
+        self.tools = [
+            TavilySearchResults(max_results=3, tavily_api_key=tavily_api_key)
+        ]
+
+        # 3. Create the Agent Prompt
+        # This tells the agent how to behave and how to use the tools.
+        prompt = ChatPromptTemplate.from_messages(
+            [
+                ("system", "You are a helpful assistant. You have access to a web search tool. Respond with the final answer to the user's question."),
+                ("placeholder", "{chat_history}"),
+                ("human", "{input}"),
+                ("placeholder", "{agent_scratchpad}"),
+            ]
+        )
+
+        # 4. Create the Agent itself
+        agent = create_tool_calling_agent(self.llm, self.tools, prompt)
+
+        # 5. Create the Agent Executor
+        # This is the runtime that will actually execute the agent's logic.
+        self.agent_executor = AgentExecutor(
+            agent=agent,
+            tools=self.tools,
+            verbose=True # Set to True to see the agent's thought process
+        )
+        print("LangChainAgent initialized.")
+
 
     def __call__(self, question: str) -> str:
         """
-        This method is called to answer a question using the Groq API.
+        This method is called to answer a question.
+        It invokes the agent executor.
         """
-        print(f"GroqAgent received question (first 50 chars): {question[:50]}...")
-
-        # A system prompt is used to guide the model to provide concise, direct answers,
-        # which is ideal for the GAIA benchmark's exact-match scoring.
-        system_prompt = (
-            "You are an expert AI agent. Your goal is to answer the following question as accurately "
-            "and concisely as possible. Provide only the final answer, without any introductory text, "
-            "explanations, or additional formatting."
-        )
-
+        print(f"LangChainAgent received question (first 50 chars): {question[:50]}...")
+        
+        # We need to handle the case where the agent makes a mistake
         try:
-            chat_completion = self.client.chat.completions.create(
-                messages=[
-                    {
-                        "role": "system",
-                        "content": system_prompt,
-                    },
-                    {
-                        "role": "user",
-                        "content": question,
-                    }
-                ],
-                model="llama3-70b-8192",  # A powerful model available via Groq
-                temperature=0.0, # Set to 0 for deterministic, factual answers
-            )
-
-            answer = chat_completion.choices[0].message.content.strip()
-            print(f"GroqAgent generated answer: {answer}")
-            return answer
-
+            response = self.agent_executor.invoke({"input": question})
+            answer = response.get("output", "No answer found.")
         except Exception as e:
-            print(f"An error occurred while calling Groq API: {e}")
-            return f"Error generating answer: {e}"
+            print(f"An error occurred in the agent executor: {e}")
+            answer = f"Agent failed with an error: {e}"
+
+        print(f"LangChainAgent generated answer: {answer}")
+        return answer
 
 
 def run_and_submit_all(profile: gr.OAuthProfile | None):
     """
-    Fetches all questions, runs the GroqAgent on them, submits all answers,
+    Fetches questions, runs the LangChainAgent on them, submits the 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
-
+    # --- Authentication and Setup ---
+    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.")
@@ -77,70 +99,45 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
     questions_url = f"{api_url}/questions"
     submit_url = f"{api_url}/submit"
 
-    # 1. Instantiate Agent (Now using GroqAgent)
+    # 1. Instantiate Agent (using the new LangChainAgent)
     try:
-        # Securely get the API key from Hugging Face Space secrets
         groq_api_key = os.getenv("GROQ_API_KEY")
-        if not groq_api_key:
-            raise ValueError("GROQ_API_KEY secret not found! Please set it in your Space's settings.")
-        agent = GroqAgent(api_key=groq_api_key)
+        tavily_api_key = os.getenv("TAVILY_API_KEY")
+        if not groq_api_key or not tavily_api_key:
+            raise ValueError("API Keys (GROQ_API_KEY, TAVILY_API_KEY) not found in secrets.")
+        agent = LangChainAgent(groq_api_key=groq_api_key, tavily_api_key=tavily_api_key)
     except Exception as e:
         print(f"Error instantiating agent: {e}")
         return f"Error initializing agent: {e}", None
 
-    # The link to your codebase (useful for verification, so please keep your space public)
     agent_code = f"https://huggingface.co/spaces/{space_id}/tree/main"
     print(f"Agent code link: {agent_code}")
 
-    # 2. Fetch Questions
+    # 2. Fetch Questions (same as before)
     print(f"Fetching questions from: {questions_url}")
     try:
-        response = requests.get(questions_url, timeout=15)
+        response = requests.get(questions_url, timeout=20)
         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
+        return f"Error fetching questions: {e}", None
 
-    # 3. Run your Agent
+    # 3. Run your Agent (same as before)
     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}"})
+        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})
 
-    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
+    # 4. Prepare Submission (same as before)
     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
+    # 5. Submit (same as before)
     print(f"Submitting {len(answers_payload)} answers to: {submit_url}")
     try:
         response = requests.post(submit_url, json=submission_data, timeout=60)
@@ -153,63 +150,28 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
             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 ---
+# --- Build Gradio Interface (Mostly the same) ---
 with gr.Blocks() as demo:
-    gr.Markdown("# Groq-Powered Agent Evaluation Runner")
+    gr.Markdown("# LangChain Agent Evaluation Runner")
     gr.Markdown(
         """
         **Instructions:**
-        1.  Make sure you have set your `GROQ_API_KEY` in the 'Secrets' section of your Space settings.
-        2.  Log in to your Hugging Face account using the button below. This is required for submission.
-        3.  Click 'Run Evaluation & Submit All Answers' to fetch questions, run your agent, submit your answers, and see your score.
-        ---
-        **Disclaimers:**
-        Once you click the "submit" button, the process can take some time as the agent answers all the questions.
-        This space provides a basic setup. You are encouraged to modify the `GroqAgent` class to experiment with different models, prompts, or even add tools to improve your score!
+        1.  Make sure you have set `GROQ_API_KEY` and `TAVILY_API_KEY` in your Space's secrets.
+        2.  Log in below. This is required for submission.
+        3.  Click 'Run Evaluation' to start the agent. You can see its thought process in the application logs!
         """
     )
-
     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)
-
-    # CORRECTED LINE: The `inputs` argument is removed. Gradio passes the
-    # OAuthProfile automatically to the `run_and_submit_all` function
-    # because of the type hint in its definition.
     run_button.click(
         fn=run_and_submit_all,
         outputs=[status_output, results_table]
@@ -217,19 +179,14 @@ with gr.Blocks() as demo:
 
 if __name__ == "__main__":
     print("\n" + "-"*30 + " App Starting " + "-"*30)
-    space_id_startup = os.getenv("SPACE_ID")
-    if space_id_startup:
-        print(f"✅ SPACE_ID found: {space_id_startup}")
-        print(f"   Repo URL: https://huggingface.co/spaces/{space_id_startup}")
-    else:
-        print("ℹ️  SPACE_ID environment variable not found (running locally?).")
-
+    # Startup checks for secrets
     if not os.getenv("GROQ_API_KEY"):
-         print("⚠️  WARNING: GROQ_API_KEY secret is not set. The app will fail if run.")
+        print("⚠️ WARNING: GROQ_API_KEY secret not set.")
     else:
-         print("✅ GROQ_API_KEY secret is set.")
-
-
+        print("✅ GROQ_API_KEY secret is set.")
+    if not os.getenv("TAVILY_API_KEY"):
+        print("⚠️ WARNING: TAVILY_API_KEY secret not set.")
+    else:
+        print("✅ TAVILY_API_KEY secret is set.")
     print("-"*(60 + len(" App Starting ")) + "\n")
-    print("Launching Gradio Interface for Groq Agent Evaluation...")
     demo.launch(debug=True, share=False)