Stripe Agent Toolkit
Give AI agents access to Stripe tools and data, and charge for AI features
- Tool type
- Agent plugin and MCP server for Stripe
- Made for
- Developers building with AI agents on Stripe
- Runs on
- MCP-compatible clients and plugin-capable agents
- Works with any MCP-compatible client; the Stripe CLI is used for setup.
What it does
Stripe Agent Toolkit gives AI agents access to Stripe developer tools, live account data, and current technical guidance. Developers install the agent plugin, which bundles the Stripe MCP server and agent skills, and the agent can then help build a Stripe integration or query metrics and financial reports from Stripe data.
A common path starts in the terminal: install the Stripe CLI, create a sandbox for API keys, then run the agent setup command. Separately, Metronome can bill for LLM token usage with usage-based metering, tiered pricing, subscriptions with metered overages, and prepaid AI credit packs, with invoices sent to Stripe for payment.
It suits developers building on Stripe with an MCP-compatible or plugin-capable agent. Agents without plugin support need to combine the MCP server and skills manually. Token billing is in public preview and requires a Metronome account, and automatic price tracking covers OpenAI, Anthropic, and Google models.
What it can do
- Gives an agent access to live Stripe account data and developer tools through the Stripe MCP server
- Installs agent skills that guide more accurate Stripe integrations
- Automatically updates bundled plugin skills and MCP server
- Provides usage-based billing for LLM tokens via Metronome, including per-token metering and markup
- Supports tiered pricing, subscriptions with metered overages, and prepaid AI credit packs
- Enables agents to query metrics, run financial reports, and analyze Stripe data
Works with
- Metronome (usage-based billing for LLM tokens, invoices sent to Stripe)
Example workflows
- Start a Stripe integration from the terminal
- Input: a developer's agent and empty project. Steps: install the Stripe CLI, run the sandbox create help command for API keys, run stripe agent setup, then ask the agent to build an integration. Output: an agent-assisted Stripe integration using sandbox keys.
- Meter LLM token usage for billing
- Input: an AI app's per-model token counts per customer. Steps: define token-based products and rates in Metronome, configure markup, send usage events with customer, model, token type and count. Output: usage charges and invoices sent to Stripe for payment.
Before you choose
- Billing for LLM tokens is in public preview and requires a Metronome account; the page advises contacting [email protected] for feedback or unsupported models.
- If your agent does not support plugins, you must combine the MCP server and agent skills yourself.
Listed under
Get started
Install the CLI, create a sandbox for API keys, then run the agent setup command.
- Install the Stripe CLI with npm install -g @stripe/cli.
- Run stripe sandbox create --help to get working API keys without signing up.
- Run stripe agent setup to install the agent plugin or skills, then ask the agent to build your integration.
Plans and prices
Billing for LLM tokens uses Metronome usage-based pricing with a markup set by the seller; no rates are stated on the source page.
We couldn’t find published prices for Stripe Agent Toolkit. See their pricing page ↗
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
- Agents and AI on Stripe | Stripe Documentation ↗docs.stripe.com
- Bill for LLM tokens | Stripe Documentation ↗docs.stripe.com
- Quickstart guides | Stripe Documentation ↗docs.stripe.com
Ratings and reviews
No ratings for Stripe Agent Toolkit yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- docs.stripe.com: product, pricing and policy pages
- YC directory mirror
- Batch and company name
- Web search
- Confirms the official site
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.