Update crewai_agent.py

crewai_agent.py

@@ -2,20 +2,23 @@ import os
 from typing import List, Dict, Any, Optional
 from dotenv import load_dotenv
 
+# Load environment variables from .env file
+load_dotenv()
+
 from langchain.agents import AgentType, initialize_agent, Tool
 from langchain.memory import ConversationBufferMemory
-from langchain.schema import BaseMessage, HumanMessage, AIMessage
+from langchain_core.messages import BaseMessage, HumanMessage, AIMessage
 from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint
 from langchain_community.tools.tavily_search import TavilySearchResults
 from langchain_community.document_loaders import WikipediaLoader, ArxivLoader
-from langchain.tools import tool
+from langchain_core.tools import tool
 from langchain.prompts import PromptTemplate
 from langchain.chains import LLMChain
 
 # Load environment variables
 GOOGLE_API_KEY = os.getenv('GOOGLE_API_KEY')
-HUGGINGFACE_API_TOKEN= os.getenv('HUGGINGFACE_API_TOKEN')
+HUGGINGFACE_API_TOKEN = os.getenv('HUGGINGFACE_API_TOKEN')
 TAVILY_API_KEY = os.getenv('TAVILY_API_KEY')
 
 
@@ -80,7 +83,10 @@ def web_search_tool(query: str) -> str:
         Formatted web search results
     """
     try:
-        search_results = TavilySearchResults(max_results=3,api_key=TAVILY_API_KEY).invoke(query)
+        if not TAVILY_API_KEY:
+            return "Error: TAVILY_API_KEY not found in environment variables"
+        
+        search_results = TavilySearchResults(max_results=3, api_key=TAVILY_API_KEY).invoke(query)
         formatted_results = "\n\n---\n\n".join([
             f'Source: {result.get("url", "")}\n\nContent:\n{result.get("content", "")}'
             for result in search_results
@@ -123,20 +129,23 @@ class LangChainAgent:
     def _get_llm(self, provider: str):
         """Get the specified LLM."""
         if provider == "google":
+            if not GOOGLE_API_KEY:
+                raise ValueError("GOOGLE_API_KEY not found in environment variables")
             return ChatGoogleGenerativeAI(
-                model="gemini-2.0-flash", 
+                model="gemini-1.5-flash", 
                 temperature=0,
                 max_tokens=2048,
-                api_key=GOOGLE_API_KEY
-            
+                google_api_key=GOOGLE_API_KEY
             )
         elif provider == "huggingface":
+            if not HUGGINGFACE_API_TOKEN:
+                raise ValueError("HUGGINGFACE_API_TOKEN not found in environment variables")
             return ChatHuggingFace(
                 llm=HuggingFaceEndpoint(
                     repo_id="microsoft/DialoGPT-medium",
                     temperature=0,
                     max_length=2048,
-                    api_key=HUGGINGFACE_API_TOKEN
+                    huggingfacehub_api_token=HUGGINGFACE_API_TOKEN
                 ),
             )
         else:
@@ -153,16 +162,21 @@ class LangChainAgent:
     
     def _create_agent(self):
         """Create the LangChain agent with tools."""
-        return initialize_agent(
-            tools=self.tools,
-            llm=self.llm,
-            agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
-            memory=self.memory,
-            verbose=True,
-            handle_parsing_errors=True,
-            max_iterations=5,
-            early_stopping_method="generate"
-        )
+        try:
+            return initialize_agent(
+                tools=self.tools,
+                llm=self.llm,
+                agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
+                memory=self.memory,
+                verbose=True,
+                handle_parsing_errors=True,
+                max_iterations=3,
+                early_stopping_method="generate"
+            )
+        except Exception as e:
+            print(f"Error creating agent: {e}")
+            # Return a simple agent without tools as fallback
+            return None
     
     def _determine_approach(self, question: str) -> str:
         """Determine the best approach for answering the question."""
@@ -190,6 +204,15 @@ class LangChainAgent:
         try:
             print(f"Processing question: {question[:100]}...")
             
+            # If agent initialization failed, use direct LLM
+            if self.agent is None:
+                print("Agent not available, using direct LLM response")
+                try:
+                    response = self.llm.invoke([HumanMessage(content=question)])
+                    return response.content
+                except Exception as llm_error:
+                    return f"Error: Unable to process question. {str(llm_error)}"
+            
             # Determine the best approach for this question
             approach = self._determine_approach(question)
             print(f"Selected approach: {approach}")
@@ -272,3 +295,4 @@ def test_langchain_agent():
 
 if __name__ == "__main__":
     test_langchain_agent()
+