Add Stripe to your agentic workflowsDeveloper preview

Use function calling to create and manage Stripe objects. For example, dynamically create Payment Links to accept funds, integrate into your support workflows to help customers, and scaffold test data.

The Stripe Agent Toolkit supports Vercel’s AI SDK, LangChain, and CrewAI. It works with any LLM provider that supports function calling and is compatible with Python and TypeScript.

import { StripeAgentToolkit } from '@stripe/agent-toolkit/ai-sdk';
import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';

const stripeAgentToolkit = new StripeAgentToolkit({
  secretKey: process.env.STRIPE_SECRET_KEY!,
  configuration: {
    actions: {
      paymentLinks: {
        create: true,
      },
      products: {
        create: true,
      },
      prices: {
        create: true,
      },
    },
  },
});

const result = await generateText({
  model: openai('gpt-4o'),
  tools: {
    ...stripeAgentToolkit.getTools(),
  },
  maxSteps: 5,
  prompt: 'Create a payment link for a new product called \"Test\" with a price of $100.',
})

Integrate usage-based billing to record usage. The Stripe Agent Toolkit offers native support for billing by prompt and completion token usage in the Vercel AI SDK. You can forward LLM costs directly to your users using a Customer and event_names for input and output Meter Events.

import { StripeAgentToolkit } from '@stripe/agent-toolkit/ai-sdk';
import { openai } from '@ai-sdk/openai';
import {
  experimental_wrapLanguageModel as wrapLanguageModel,
} from 'ai';

const model = wrapLanguageModel({
  model: openai('gpt-4o'),
  middleware: stripeAgentToolkit.middleware({
    billing: {
      customer: process.env.STRIPE_CUSTOMER_ID!,
      meters: {
        input: process.env.STRIPE_METER_INPUT!,
        output: process.env.STRIPE_METER_OUTPUT!,
      },
    },
  }),
});

Usage-based billing gives you the flexibility to charge for any unit, such as by time. You can emit billing events directly, such as measuring duration:

import stripe
import time

start = time.time()

# Run your agent workflow, LLM execution, etc.
execute_agent()

end = time.time()

stripe.billing.MeterEvent.create(
  event_name="agent_execution",
  payload={
    "value": end - start,
    "stripe_customer_id": "{{CUSTOMER_ID}}"
  },
)

Use Stripe Issuing to create single-use virtual cards for your business purchases. This allows your agents to spend funds. The Issuing APIs allow you to programmatically approve or decline authorizations, ensuring your purchase intent matches the authorization. Spending controls allow you to set budgets and limit spending for your agents.