mcpField
S
stripekit

stripekit

Local
github.com/rafaelcgFinance2Updated 12w ago

About stripekit

stripekit is a CLI for managing a Stripe catalog from repository configuration and setting up billing code, for developers and coding agents. It plans, pushes, pulls, and checks account state, and scaffolds billing code for Next.js App Router projects.

Key features

  1. 01Manage Stripe catalog declaratively
  2. 02Plan, push, and pull config
  3. 03Check account and config drift
  4. 04Generate Next.js billing code
  5. 05Forward Stripe webhooks locally

Use cases

  1. 01Sync repository config to a Stripe account
  2. 02Set up checkout for a Next.js app
  3. 03Check keys, webhooks, and portal wiring

MCP support

`stripekit mcp` provides an MCP server exposing `plan`, `push`, `pull`, and `check` as tools for agents.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"rafaelcg-stripekit": {
"command": "npx",
"args": [
"-y",
"stripekit"
]
}
}
}
Run commandnpx -y stripekit

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmstripekit0.2.2

Tools

4

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
cwdstringNoProject directory containing stripe.config.ts (defaults to the server working directory).
urlstringNoBase URL used when resolving the webhook endpoint.

Details

From the project README

stripekit

npm CI license docs

The create-next-app of Stripe. One command takes an app from zero to working payments on your Stripe account: it provisions your catalog from a config file in your repo, and drops in correct-by-construction billing code you own. No hosted service. No revenue share. Nothing to depend on at runtime. Works identically whether a human or an agent runs it.

Documentation →

Repository layout

PathWhat it is
packages/stripekitThe stripekit npm package — CLI + config API.
examples/playgroundA scratch project to try stripekit against a Stripe test account.
docsThe VitePress documentation site.
DESIGN.mdThe MVP design: CLI surface, config schema, generated code, build order.

Status

Built and tested: stripekit init | plan | push | pull | dev | check | mcp.

  • Reconciler (plan / push / pull) — reconciles your Stripe account to a declarative stripe.config.ts (products, prices, webhook endpoint, customer portal), treating prices as immutable, archiving instead of deleting, and converging idempotently.
  • init — scaffolds stripe.config.ts plus correct-by-construction billing code (webhook handler, state sync, checkout/portal routes) into a Next.js App Router app.
  • dev — wraps stripe listen to forward webhooks locally and capture the signing secret.
  • check — a doctor that verifies keys, config validity, config-vs-account drift, and webhook/portal wiring.

There's also a VitePress docs site with a landing page — run pnpm docs:dev. Roadmap (see DESIGN.md): init support beyond Next.js and richer catalog features.

See the package README for usage.

For AI agents

stripekit is built to be driven by agents, not just humans:

  • llms.txt and llms-full.txt — the docs as clean markdown for LLM context.
  • SKILL.md — point your coding agent at this and it can set stripekit up for you.
  • stripekit mcp — an MCP server exposing plan/push/pull/check as tools an agent can call.
  • Every command accepts --json and --yes for deterministic, non-interactive use.

Install the MCP server one-click in Cursor:

Add to Cursor

…or in Claude Code as a plugin:

/plugin marketplace add rafaelcg/stripekit
/plugin install stripekit@stripekit-marketplace

See the MCP docs for other clients (Claude Desktop, .cursor/mcp.json, claude mcp add).

Working on stripekit? See AGENTS.md.

Develop

pnpm install
pnpm -C packages/stripekit build      # bundle CLI + types
pnpm -C packages/stripekit test       # unit + integration tests
pnpm -C packages/stripekit typecheck

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.