LangChain Tools Explained for Beginners: Add Function Calling to Your AI Agent

📘 Introduction

A basic LLM can generate text, but an AI agent often needs to do more than write an answer. It may need to calculate a price, look up product information, check a database, call an API, or run a small business rule. LangChain tools make this possible by exposing Python functions to a chat model in a structured way.

In this tutorial, you will learn what LangChain tools are, why tool calling matters for AI agents, and how to define a simple custom tool in Python. Then we will connect that tool to a chat model so the model can decide when to call it.

💡 What are LangChain tools?

A LangChain tool is a callable function with a name, description, and input schema. The description helps the model understand when the tool should be used. The input schema helps the model provide the right arguments.

You can think of a normal Python function as something your code calls directly. A LangChain tool is a Python function prepared so a model can request it during a conversation.

🧠 Why tools matter for AI agents

Tools make AI agents more useful because they connect language understanding with real operations. A model can understand the user request, choose the right tool, provide structured inputs, and use the tool result to continue the conversation.

User request -> Model decides tool call -> Python tool runs -> Tool result returns -> Model answers

✅ Prerequisites

Before getting started, make sure you have:

☑️ Python installed
☑️ Basic Python knowledge
☑️ An OpenAI API key or another tool-calling chat model provider
☑️ A terminal or command prompt

⚙️1️⃣ Create a project folder

Create a new local project folder for this tutorial:

mkdir langchain-tools-agent
cd langchain-tools-agent

🐍2️⃣ Create a virtual environment

Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate

On Windows, activate it with:

.venv\Scripts\activate

📦3️⃣ Install libraries

Install the packages we need for tools and the chat model integration:

pip install -U langchain langchain-openai

🔐4️⃣ Set your API key

Set your OpenAI API key as an environment variable. Replace the placeholder with your own key and never commit real secrets to a repository.

export OPENAI_API_KEY="your_api_key_here"

On Windows PowerShell, use:

$env:OPENAI_API_KEY="your_api_key_here"

🛠️5️⃣ Define your first LangChain tool

Create a file named define_tool.py. In this public example, we define a small discount calculator tool and call it directly so you can understand the tool object before adding a model.

from langchain.tools import tool


@tool
def calculate_discount(price: float, discount_percent: float) -> str:
    """Calculate the final price after a percentage discount."""
    if price < 0:
        return "Price must be greater than or equal to 0."
    if discount_percent < 0 or discount_percent > 100:
        return "Discount percent must be between 0 and 100."

    final_price = price * (1 - discount_percent / 100)
    return f"Final price: ${final_price:.2f}"


print(calculate_discount.name)
print(calculate_discount.description)
print(calculate_discount.invoke({"price": 120, "discount_percent": 25}))

Run the script:

python define_tool.py

You should see the tool name, description, and the calculated result:

calculate_discount
Calculate the final price after a percentage discount.
Final price: $90.00