LangChain Agents: Conversational React Description

Question

Previously I used initialize_agent method by passing agent=AgentType.CONVERSATIONAL_REACT_DESCRIPTION to initialize a conversation react agent in LangChain v0.0.

agent_executor = initialize_agent(tools=tools, 
                                  llm=llm, 
                                  memory=memory, 
                                  verbose=True, 
                                  max_iterations=3, 
                                  handle_parsing_errors=True, 
                                  agent=AgentType.CONVERSATIONAL_REACT_DESCRIPTION, 
                                  agent_kwargs=prompt
                                  )

But in LangChain v1.0, it says that it will be deprecated and to use the create_react_agent method. I want to initialize an CONVERSATIONAL_REACT_DESCRIPTION agent.

# Construct the ReAct agent
agent = create_react_agent(llm=llm, 
                           tools=tools, 
                           prompt=template
                           )
# Create an agent executor by passing in the agent and tools
agent_executor = AgentExecutor(agent=agent, 
                               tools=tools, 
                               max_iterations=3,
                               handle_parsing_errors=True,
                               verbose=True
                              )

langchain 0.1.11, langchain-community 0.0.27, langchain-core 0.1.30, langchain-google-genai 0.0.9, langchain-mistralai 0.0.5, langchain-openai 0.0.8, langchain-text-splitters 0.0.1, langsmith 0.1.23

How to initiate a CONVERSATIONAL_REACT_DESCRIPTION in LangChain v0.1?

Answer 1

import os
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
from langchain_core.prompts import MessagesPlaceholder
from langchain_openai import ChatOpenAI
from langchain.agents import create_openai_functions_agent
from langchain.agents import AgentExecutor
from langchain.tools.retriever import create_retriever_tool

vectordb = FAISS.load_local(faiss_path, instructor_embeddings)
retriever = vectordb.as_retriever(score_threshold=0.7)

retriever_tool = create_retriever_tool(
    retriever,
    "some_name",
    "Retriver description",
)

prompt = ChatPromptTemplate.from_messages([
    ("system", prompt_template),
    MessagesPlaceholder(variable_name="chat_history"),
    ("user", "{input}"),
    MessagesPlaceholder(variable_name="agent_scratchpad"),
])
tools = [retriever_tool, <other_tools>]

llm = ChatOpenAI(temperature=0.0)
agent = create_openai_functions_agent(llm, tools, prompt)
agent_executor = AgentExecutor(agent=agent, tools=tools)

data = {
        "chat_history": <chat_history>,
        "input": "<user prompt>"
       }
response = agent_executor.invoke(data)
response = response['output']