first agent defintion

app.py

@@ -3,21 +3,56 @@ import gradio as gr
 import requests
 import inspect
 import pandas as pd
+from langchain.agents import initialize_agent, Tool
+from langchain.agents.agent_types import AgentType
+from langchain.chat_models import ChatOpenAI
+from langchain.tools import DuckDuckGoSearchRun
 
 # (Keep Constants as is)
 # --- Constants ---
 DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
+# 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
+
 # --- Basic Agent Definition ---
 # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
-class BasicAgent:
+class BasicAgent: # Some times Inheritance is needed
     def __init__(self):
-        print("BasicAgent initialized.")
+        print("BasicAgent with LangChain initialized.")
+
+        # Create the LLM # Temprature set to 0 because we need exact match
+        llm = ChatOpenAI(temperature=0, model="gpt-3.5-turbo")  # ChatML understands roles (user, assistant, system)
+
+        # Define tools # Description matters a lot, unless youre using @tool as a decorator and a python function (pulls docstring as tool description)
+        search = DuckDuckGoSearchRun()
+        tools = [
+            Tool(name="Search", 
+                 func=search.run, 
+                 description="Useful for answering questions about current events, facts, or general knowledge by querying the web."),
+            # Can add custom Gaia tools here later
+        ]
+
+        # Initialize agent # Langchain automatically creates a system prompt from this initialization depending on its parameters
+        self.agent = initialize_agent(
+            tools, # List of tools with names + descriptions
+            llm, # The LLM to generate actions/thoughts
+            agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION, # Agent behavior type, in this case set to ReAct-style agent (great choice but depends)
+            verbose=True # Prints out each reasoning step (for debugging)
+        )
+    # Function to be called by the workflow
     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
+        # Do not return intermidiate steps or thoughts
+        response = self.agent.run(question) 
+        print(f"Agent response: {response}")
+        return response
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """