Update agent.py

agent.py

@@ -2,100 +2,50 @@ import os
 from dotenv import load_dotenv
 from typing import List, Dict, Any, Optional
 import tempfile
-import requests
-from urllib.parse import urlparse
 import pandas as pd
 import numpy as np
 
 """Langraph"""
 from langgraph.graph import START, StateGraph, MessagesState
-from langchain_community.tools import DuckDuckGoSearchResults
-from langchain_community.document_loaders import WikipediaLoader
-from langchain_community.document_loaders import ArxivLoader
 from langgraph.prebuilt import ToolNode, tools_condition
 from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_groq import ChatGroq
 from langchain_huggingface import (
     ChatHuggingFace,
     HuggingFaceEndpoint,
-    HuggingFaceEmbeddings,
 )
-from langchain_community.vectorstores import SupabaseVectorStore
 from langchain_core.messages import SystemMessage, HumanMessage
 from langchain_core.tools import tool
-from langchain.tools.retriever import create_retriever_tool
-from supabase.client import Client, create_client
 from langchain_openai import ChatOpenAI
+from langchain_community.utilities import SerpAPIWrapper
 
 load_dotenv()
 
 ### =============== SEARCH TOOLS =============== ###
 
 @tool
-def web_search(query: str) -> str:
-    """Search DuckDuckGo for a query and return maximum 3 results.
+def serpapi_search(query: str) -> str:
+    """Search the web using SerpAPI.
     Args:
         query: The search query."""
     try:
-        search_results = DuckDuckGoSearchResults(max_results=3).invoke(query=query)
-        # DuckDuckGo returns results as a string, so we need to parse it
-        # The format is typically [snippet: text, title: title, link: url]
-        formatted_results = []
+        # Get API key from environment variable
+        api_key = os.getenv("SERPAPI_API_KEY")
+        if not api_key:
+            return {"search_results": "Error: SERPAPI_API_KEY not found in environment variables."}
         
-        # Remove outer quotes if present and format properly
-        if isinstance(search_results, str):
-            # Simple parsing of the DuckDuckGo result format
-            result_entries = search_results.split('], [')
-            for entry in result_entries:
-                # Clean up the entry
-                entry = entry.replace('[', '').replace(']', '').strip()
-                if entry:
-                    # Extract components
-                    parts = {}
-                    for part in entry.split(', '):
-                        if ': ' in part:
-                            key, value = part.split(': ', 1)
-                            parts[key] = value
-                    
-                    # Format as document
-                    if 'snippet' in parts and 'link' in parts:
-                        doc_content = parts.get('snippet', '')
-                        doc_source = parts.get('link', '')
-                        formatted_results.append(
-                            f'<Document source="{doc_source}" page=""/>\n{doc_content}\n</Document>'
-                        )
+        # Initialize SerpAPIWrapper with the API key
+        search = SerpAPIWrapper(serpapi_api_key=api_key)
         
-        if not formatted_results:
-            return {"web_results": "No results found for the query. Try using wikipedia_search instead."}
+        # Perform the search
+        results = search.run(query)
         
-        return {"web_results": "\n\n---\n\n".join(formatted_results)}
-    except Exception as e:
-        return {"web_results": f"Error performing web search: {str(e)}. Try using wikipedia_search instead."}
-
-@tool
-def wikipedia_search(query: str) -> str:
-    """Search Wikipedia for a query.
-    Args:
-        query: The search query."""
-    try:
-        # Use WikipediaLoader which works better in restricted environments
-        docs = WikipediaLoader(query=query, load_max_docs=2).load()
-        
-        if not docs:
-            return {"wikipedia_results": "No Wikipedia results found for the query."}
-        
-        formatted_results = []
-        for doc in docs:
-            title = doc.metadata.get('title', 'Unknown Title')
-            source = doc.metadata.get('source', 'Wikipedia')
-            content = doc.page_content[:1000] + "..." if len(doc.page_content) > 1000 else doc.page_content
-            formatted_results.append(
-                f'<Document source="{source}" title="{title}">\n{content}\n</Document>'
-            )
+        if not results or results.strip() == "":
+            return {"search_results": "No search results found."}
         
-        return {"wikipedia_results": "\n\n---\n\n".join(formatted_results)}
+        return {"search_results": results}
     except Exception as e:
-        return {"wikipedia_results": f"Error searching Wikipedia: {str(e)}."}
+        return {"search_results": f"Error performing search: {str(e)}"}
 
 ### =============== DOCUMENT PROCESSING TOOLS =============== ###
 # File handling still requires external tools
@@ -130,8 +80,7 @@ print(system_prompt)
 sys_msg = SystemMessage(content=system_prompt)
 
 tools = [
-    web_search,
-    wikipedia_search,
+    serpapi_search,
     save_and_read_file,
 ]
 
@@ -141,7 +90,7 @@ def build_graph(provider: str = "openai"):
     """Build the graph"""
     # Load environment variables from .env file
     if provider == "openai":
-        llm = ChatOpenAI(model="gpt-4.1", temperature=0)
+        llm = ChatOpenAI(model="gpt-4o", temperature=0)
     elif provider == "groq":
         llm = ChatGroq(model="qwen-qwq-32b", temperature=0)
     elif provider == "huggingface":