1. Agent & Tools Basics

In this section, you will build a basic ReAct (Reasoning and Acting) agent from scratch and expose a simple python function as a tool to retrieve the current system time.

Objectives

Expose a Python function as a custom LangChain tool using the base Tool class.
Load prompt templates directly from the LangChain prompt hub.
Configure create_react_agent and handle agent execution loops using AgentExecutor.

Implementation Plan

Goal

Create a ReAct agent with access to a local time tool, query it for the current time, and print the execution trajectory.

Sample Input

{"input": "What time is it?"}

Sample Output

An executor output dictionary returning the current formatted time.

Plan

Define a Python helper function get_current_time that returns system time in H:MM AM/PM format.
Instantiate a custom tool using the Tool(...) constructor mapping to the helper function.
Pull the standard ReAct prompt hwchase17/react from the LangChain prompt hub.
Initialize the chat model (ChatOpenAI) and create the ReAct agent using create_react_agent.
Create an AgentExecutor with verbose=True and invoke it with the input question.

Step-by-Step Implementation

Step 1: Define the Python Tool Function

We create the base function that our tool will execute when triggered by the agent.

def get_current_time(*args, **kwargs):
    """Returns the current time in H:MM AM/PM format."""
    import datetime
    now = datetime.datetime.now()
    return now.strftime("%I:%M %p")

Step 2: Instantiate the Tool

We wrap our Python function in the Tool class, specifying a descriptive name and description to help the LLM understand when it is appropriate to use this tool.

from langchain_core.tools import Tool

tools = [
    Tool(
        name="Time",
        func=get_current_time,
        description="Useful for when you need to know the current time",
    ),
]

Step 3: Pull Prompt and Create Agent

We pull the standard ReAct prompt from the hub and pass it along with the model and tools to initialize the agent.

from langchain import hub
from langchain.agents import create_react_agent
from langchain.chat_models import init_chat_model

prompt = hub.pull("hwchase17/react")
llm = init_chat_model("llama-3.3-70b-versatile", model_provider="groq", temperature=0)

agent = create_react_agent(
    llm=llm,
    tools=tools,
    prompt=prompt,
    stop_sequence=True,
)

Step 4: Run Agent Executor

We create the executor loop to handle step logs, parsing errors, and final response retrieval.

from langchain.agents import AgentExecutor

agent_executor = AgentExecutor.from_agent_and_tools(
    agent=agent,
    tools=tools,
    verbose=True,
)

response = agent_executor.invoke({"input": "What time is it?"})
print("response:", response)

Complete Combined Code

Below is the complete, consolidated Python script uniting all of the steps above:

from dotenv import load_dotenv
from langchain import hub
from langchain.agents import (
    AgentExecutor,
    create_react_agent,
)
from langchain_core.tools import Tool
from langchain.chat_models import init_chat_model

# Load environment variables from .env file
load_dotenv()


# Define a very simple tool function that returns the current time
def get_current_time(*args, **kwargs):
    """Returns the current time in H:MM AM/PM format."""
    import datetime  # Import datetime module to get current time

    now = datetime.datetime.now()  # Get current time
    return now.strftime("%I:%M %p")  # Format time in H:MM AM/PM format


# List of tools available to the agent
tools = [
    Tool(
        name="Time",  # Name of the tool
        func=get_current_time,  # Function that the tool will execute
        # Description of the tool
        description="Useful for when you need to know the current time",
    ),
]

# Pull the prompt template from the hub
prompt = hub.pull("hwchase17/react")

# Initialize a Groq model using core abstractions
llm = init_chat_model(
    "llama-3.3-70b-versatile", model_provider="groq", temperature=0
)

# Create the ReAct agent using the create_react_agent function
agent = create_react_agent(
    llm=llm,
    tools=tools,
    prompt=prompt,
    stop_sequence=True,
)

# Create an agent executor from the agent and tools
agent_executor = AgentExecutor.from_agent_and_tools(
    agent=agent,
    tools=tools,
    verbose=True,
)

# Run the agent with a test query
response = agent_executor.invoke({"input": "What time is it?"})

# Print the response from the agent
print("response:", response)

Practice & Exercises

To practice setting up basic agents, open the interactive notebook:

Practice & Exercises

Practice initializing ReAct agents and wrapping custom functions in Tool objects.💻 VS Code | 🚀 Colab | 📥 Download

Introduction

Chat Models

Templates & Parsers

Chains

RAG

Tools and Agents

Objectives

Implementation Plan

Goal

Sample Input

Sample Output

Plan

Step-by-Step Implementation

Step 1: Define the Python Tool Function

Step 2: Instantiate the Tool

Step 3: Pull Prompt and Create Agent

Step 4: Run Agent Executor

Complete Combined Code

Practice & Exercises

Practice & Exercises

​Objectives

​Implementation Plan

​Goal

​Sample Input

​Sample Output

​Plan

​Step-by-Step Implementation

​Step 1: Define the Python Tool Function

​Step 2: Instantiate the Tool

​Step 3: Pull Prompt and Create Agent

​Step 4: Run Agent Executor

​Complete Combined Code

​Practice & Exercises

Practice & Exercises

Objectives

Implementation Plan

Goal

Sample Input

Sample Output

Plan

Step-by-Step Implementation

Step 1: Define the Python Tool Function

Step 2: Instantiate the Tool

Step 3: Pull Prompt and Create Agent

Step 4: Run Agent Executor

Complete Combined Code

Practice & Exercises