ReAct Agent Documentation

Functions

create_agent(llm, tools, callbacks=[])

Creates and initializes a LangChain agent configured with the ZERO_SHOT_REACT_DESCRIPTION strategy.

This agent uses provided tools and a large language model (LLM) to perform tasks using a zero-shot reasoning and action (ReAct) methodology.

Parameters:

Name	Type	Description	Default
llm		The large language model instance used by the agent for generating responses.	required
tools	list	A list of Tool objects that the agent can use to accomplish tasks.	required
callbacks	list	A list of callback functions to monitor agent interactions. Defaults to an empty list.	[]

Returns:

Type	Description
	An initialized LangChain agent ready to perform ReAct-style tasks.

Example

from langchain.llms import OpenAI
from langchain.agents import Tool
from my_tools import custom_tool

llm = OpenAI(model="gpt-3.5-turbo")
tools = [custom_tool]

agent = create_agent(llm, tools)

response = agent.run("Extract invoice data from the provided document.")
print(response)