StripeAgentToolkit integration | LangChain English

This guide provides a quick overview for getting started with Stripe's agent toolkit.

You can read more about StripeAgentToolkit in Stripe's launch blog or on the project's PyPI page.

Overview

Integration details

Class	Package	Serializable	JS Support	Version
`StripeAgentToolkit`	`stripe-agent-toolkit`	❌	✅

Setup

This externally-managed package is hosted out of the stripe-agent-toolkit project, which is managed by Stripe's team.

You can install it, along with langgraph for the following examples, with pip:

pip install --quiet -U langgraph stripe-agent-toolkit

Credentials

In addition to installing the package, you will need to configure the integration with your Stripe account's secret key, which is available in your Stripe Dashboard.

import getpass
import os

if not os.environ.get("STRIPE_SECRET_KEY"):
    os.environ["STRIPE_SECRET_KEY"] = getpass.getpass("STRIPE API key:\n")

It's also helpful (but not needed) to set up LangSmith for best-in-class observability:

os.environ["LANGSMITH_TRACING"] = "true"
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass()

Instantiation

Here we show how to create an instance of the Stripe Toolkit

from stripe_agent_toolkit.langchain.toolkit import StripeAgentToolkit

stripe_agent_toolkit = StripeAgentToolkit(
    secret_key=os.getenv("STRIPE_SECRET_KEY"),
    configuration={
        "actions": {
            "payment_links": {
                "create": True,
            },
        }
    },
)

Agent

Here's how to use the toolkit to create a basic agent in langgraph:

from langchain_anthropic import ChatAnthropic
from langchain.agents import create_agent


model = ChatAnthropic(
    model="claude-sonnet-4-6",
)

langgraph_agent_executor = create_agent(model, stripe_agent_toolkit.get_tools())

input_state = {
    "messages": """
        Create a payment link for a new product called 'test' with a price
        of $100. Come up with a funny description about buy bots,
        maybe a haiku.
    """,
}

output_state = langgraph_agent_executor.invoke(input_state)

print(output_state["messages"][-1].content)