Add chat history support

app.py

@@ -17,6 +17,7 @@ from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain.schema import SystemMessage
 import logging
 import sys
+from langchain_community.utilities import DuckDuckGoSearchAPIWrapper
 
 # LangChain
 from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, ToolMessage
@@ -49,8 +50,14 @@ app = FastAPI()
 def search_web(query: str) -> str:
     """Search the web for information on a given query."""
     logger.info(f"Searching web for query: {query}")
-    # In a real implementation, this would connect to a search API
-    return f"Found information about {query}: This is a simulated web search result for '{query}'."
+    try:
+        search = DuckDuckGoSearchAPIWrapper()
+        results = search.run(query)
+        logger.info(f"Search results: {results}")
+        return results
+    except Exception as e:
+        logger.error(f"Error in web search: {str(e)}")
+        return f"Error searching the web: {str(e)}"
 
 
 @tool
@@ -143,18 +150,29 @@ Use these tools to assist the user with their requests. When a tool is needed, c
     return prompt | model.bind_tools(tools=tools)
 
 
-# WebSocket for real-time communication
+# WebSocket endpoint for real-time communication
 @app.websocket("/ws")
 async def websocket_endpoint(websocket: WebSocket):
     logger.info("New WebSocket connection established")
     await websocket.accept()
+    chat_history = []  # Initialize chat history
     try:
         while True:
             data = await websocket.receive_text()
             logger.info(f"Received message: {data}")
             try:
-                # Process the message with the agent
-                response = agent_executor.invoke({"input": data, "chat_history": []})
+                # Add user message to chat history
+                chat_history.append(HumanMessage(content=data))
+                
+                # Process the message with the agent, including chat history
+                response = agent_executor.invoke({
+                    "input": data,
+                    "chat_history": chat_history
+                })
+                
+                # Add AI response to chat history
+                chat_history.append(AIMessage(content=response["output"]))
+                
                 logger.info(f"Agent response: {response['output']}")
                 await websocket.send_json({"type": "ai_message", "content": response["output"]})
             except Exception as e: