Stripe Skill logo

Stripe Skill

Community
zeenie-ai
stripe-skill

Process payments, manage customers, issue refunds, and handle subscriptions via the Stripe CLI. Pass any Stripe command (customers, charges, payment_intents, refunds, invoices, products, prices, subscriptions) and the tool runs it for you and returns parsed JSON.

Overview

Publisherzeenie-ai
RepositoryOpenCompany
Skill namestripe-skill
Stars
912
Forks
137
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by zeenie-ai on GitHub. Read the source before you install it.

Installation

Install the Stripe Skill AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/zeenie-ai/OpenCompany.git /tmp/OpenCompany
mkdir -p .claude/skills
cp -r /tmp/OpenCompany/server/skills/payments_agent/stripe-skill .claude/skills/stripe-skill
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Stripe Skill in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Stripe Skill on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Stripe Skill is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Stripe Skill

Pass-through over the official Stripe CLI. The tool runs stripe <command> --api-key <stored-key> for you and returns parsed JSON. Every Stripe resource the CLI supports works without code changes — products and prices created tomorrow work the same way.

Tool: stripe_action

Single field: command — a string identical to what you would type after stripe on the terminal.

Schema

FieldTypeRequiredDescription
commandstringYesStripe CLI command (e.g. "customers create --email a@b.com"). Quote arguments that contain spaces.

Response

json
{
  "command": "customers create --email a@b.com",
  "success": true,
  "result": {
    "id": "cus_NkXz123",
    "email": "a@b.com",
    "object": "customer",
    "created": 1730000000,
    ...
  },
  "stdout": "<raw CLI stdout>"
}

On CLI failure the tool raises an error with Stripe's own message (e.g. "No such customer: cus_invalid").

Common commands

Customers

CommandPurpose
customers create --email <addr>Create a customer. Optional: --name, --description, --phone, --metadata key=val.
customers retrieve <cus_id>Fetch a customer by id.
customers list --limit 10List recent customers. Optional: --email <addr> filter.
customers update <cus_id> --email <new>Update fields on a customer.
customers delete <cus_id>Permanently delete a customer.

Search:

CommandPurpose
customers search --query "email:'a@b.com'"Server-side search across customers. Stripe Search query syntax.

PaymentIntents (recommended for charges)

CommandPurpose
payment_intents create --amount 2000 --currency usdCreate a PaymentIntent for $20.00. Amount is in the smallest currency unit (cents for USD).
payment_intents create --amount 2000 --currency usd --customer cus_123Create attached to a saved customer.
payment_intents create --amount 2000 --currency usd --confirm --payment-method pm_card_visaCreate + confirm in one shot (most common test path).
payment_intents retrieve pi_123Fetch by id.
payment_intents list --limit 20List recent intents.
payment_intents cancel pi_123Cancel an unconfirmed or processing intent.

Charges (legacy direct-charge path)

CommandPurpose
charges retrieve ch_123Fetch a charge by id.
charges list --limit 10List recent charges.
charges list --customer cus_123List charges for a specific customer.

For new integrations prefer PaymentIntents over direct charges create — Stripe deprecated raw card-token charge creation.

Refunds

CommandPurpose
refunds create --payment-intent pi_123Refund a PaymentIntent in full.
refunds create --payment-intent pi_123 --amount 500Partial refund (in smallest currency unit).
refunds create --charge ch_123Refund by charge id (older flow).
refunds retrieve re_123Fetch a refund by id.
refunds list --limit 10List recent refunds.

Invoices & Subscriptions

CommandPurpose
invoices create --customer cus_123Draft an invoice.
invoices finalize_invoice in_123Finalize a draft (locks for sending/charging).
invoices pay in_123Charge the customer for a finalized invoice.
invoices list --customer cus_123List invoices for a customer.
subscriptions create --customer cus_123 --items='[{"price":"price_xxx"}]'Create a subscription.
subscriptions retrieve sub_123Fetch a subscription.
subscriptions cancel sub_123Cancel immediately.
subscriptions update sub_123 --cancel-at-period-end trueCancel at end of current billing period.

Products & Prices

CommandPurpose
products create --name "Pro Plan"Create a product.
prices create --product prod_123 --unit-amount 2000 --currency usd --recurring='{"interval":"month"}'Create a recurring price tied to a product.
prices list --product prod_123List prices for a product.

Trigger (synthetic test events)

CommandPurpose
trigger payment_intent.succeededFire a synthetic PaymentIntent succeeded event.
trigger charge.refundedFire a synthetic refund event.
trigger customer.subscription.createdFire a synthetic subscription event.

stripe trigger is invaluable for testing webhook flows — it sends the event to Stripe, which sends it back through the same stripe listen daemon and into your stripeReceive trigger node.

Common workflows

Create a customer and charge them

json
{ "command": "customers create --email rosy@sparrow.com --name 'Rosy Sparrow'" }

returns cus_…. Then:

json
{ "command": "payment_intents create --amount 2000 --currency usd --customer cus_xxx --payment-method pm_card_visa --confirm" }

Refund the most recent payment for a customer

json
{ "command": "payment_intents list --customer cus_123 --limit 1" }

read data[0].id, then:

json
{ "command": "refunds create --payment-intent pi_xxx" }

Set up a recurring subscription

json
{ "command": "products create --name 'Pro Plan'" }     -> prod_xxx
{ "command": "prices create --product prod_xxx --unit-amount 2000 --currency usd --recurring='{\"interval\":\"month\"}'" }   -> price_xxx
{ "command": "subscriptions create --customer cus_xxx --items='[{\"price\":\"price_xxx\"}]'" }

Quoting and escaping

The command string is parsed with shlex.split before being handed to stripe. Use single quotes for arguments that contain spaces:

You want to sendType this in command
--name=Acme Inccustomers create --name 'Acme Inc'
metadata with spacescustomers update cus_x --metadata 'plan=Acme Pro'
a JSON array as a flag valuesubscriptions create --items='[{"price":"price_xxx"}]'

JSON inside --items / --recurring etc. should be a single quoted JSON string with no internal spaces — the CLI parses it on the server side.

Idempotency

For any create/charge that you don't want to duplicate on retry, add an idempotency key:

customers create --email a@b.com -H "Idempotency-Key: order_42_user_7"

The CLI accepts custom headers via -H. Stripe deduplicates by the key for 24 hours.

Test mode vs live mode

The stored API key (set in the Credentials Modal) determines mode:

Key prefixModeEffect
sk_test_…TestFake money, fake events, no real charges. Always use this for development.
sk_live_…LiveReal money. Real customers. Use only after thorough testing.
rk_test_… / rk_live_…RestrictedScoped permissions; recommended for production agents that don't need full account access.

The CLI reports livemode: true/false on every response object so you can verify which environment a result is from.

Errors

Stripe returns structured errors that surface in the tool's error field. Common ones:

Error codeMeaningRecovery
resource_missingThe id you passed doesn't existVerify the id; list to find the right one
card_declinedCard payment failedUse a different --payment-method; in test mode try pm_card_visa
invalid_request_errorWrong / missing argumentCheck stripe <resource> create --help
authentication_required3DS challenge neededUse a setup_intent or off-session payment_method
rate_limit_errorToo many requestsPause and retry; Stripe rate limit is 100/sec in live, 25/sec in test

The CLI's stderr always contains the Stripe request id (req_…) — include it when reporting issues.

Webhooks (stripeReceive trigger)

When the Stripe daemon is connected (Credentials Modal → Stripe → Connect), every event Stripe sends is forwarded to OpenCompany and delivered to any active stripeReceive trigger nodes after HMAC-SHA256 signature verification (Stripe-Signature header).

To test event delivery without making real payments, use the trigger command above. Events fire through the same path as real ones.

Filters on the trigger node (event_type_filter):

FilterMatches
allevery event
charge.succeededexact match
charge.*every charge.* event (succeeded, refunded, failed, …)
payment_intent.*every PaymentIntent event

The trigger output mirrors Stripe's event shape: event_id, event_type, created, livemode, api_version, request_id, account, data.

Authentication — stripe login (browser OAuth)

Authentication is delegated entirely to the Stripe CLI. There is no API key to paste into OpenCompany. The Credentials Modal's Login with Stripe button drives the CLI's two-step machine-friendly login:

  1. Browser-side: the modal opens https://dashboard.stripe.com/stripecli/auth/... in a new tab and shows a verification code. The user signs in to Stripe and confirms the code.
  2. CLI-side: the CLI polls Stripe until the user authorises, then writes credentials (one restricted key per mode, valid for 90 days) to ~/.config/stripe/config.toml (or $XDG_CONFIG_HOME/stripe/config.toml).
  3. OpenCompany-side: when login completes, the listen daemon starts automatically and captures the webhook signing secret.

After login, every stripe … command run via this tool reads its credentials from the CLI's config file — no --api-key flag in any command you'd type.

Logout (Credentials Modal → Disconnect) stops the listen daemon and runs stripe logout --all to clear the config file.

Best practices

  1. Restricted keys are auto-issued. When you stripe login, the CLI generates restricted keys with CLI-appropriate scopes (one per test/live mode) — you don't pick scopes manually.
  2. Default to test mode for development. The CLI distinguishes modes; livemode: true/false is on every response.
  3. Add idempotency keys to all create operations. Especially for charges and refunds — accidental retry without a key duplicates the charge.
  4. Quote string arguments with spaces (--name 'Acme Inc').
  5. Use PaymentIntents, not raw charges for new integrations.
  6. Surface Stripe error messages verbatim to the user. They are precise and actionable; don't paraphrase them.
  7. Don't include --api-key in your command string. The CLI already has stored credentials; an inline --api-key overrides them and risks key leakage in process listings and logs.

Setup checklist

  1. Stripe CLI binary — auto-downloaded on first Login with Stripe click if not already on PATH. Resolution order:
    • System install (brew install stripe/stripe-cli/stripe, scoop install stripe, apt install stripe, or a direct binary from https://stripe.com/docs/stripe-cli#install).
    • OpenCompany package cache at <DATA_DIR>/packages/stripe/bin/stripe[.exe] (populated automatically from GitHub releases — pinned to a known-good CLI version).
  2. Credentials Modal → Stripe → Login with Stripe → a browser tab opens to the Stripe Dashboard with a pairing code. Authorise. The modal flips to "Connected" when the CLI's login --complete subprocess returns and the listen daemon spins up (webhook_secret_captured: true).
  3. The Stripe action node is connected to your agent's input-tools handle.
  4. (For webhook flows) A stripeReceive trigger node is wired into your workflow with the event-type filter you care about.

Frequently asked questions

What does the Stripe Skill AI skill do?

Process payments, manage customers, issue refunds, and handle subscriptions via the Stripe CLI. Pass any Stripe command (customers, charges, payment_intents, refunds, invoices, products, prices, subscriptions) and the tool runs it for you and returns parsed JSON.

Why use Stripe Skill on TypingMind?

Because you install it once and use it with any model. Stripe Skill is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Stripe Skill in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/zeenie-ai/OpenCompany/tree/main/server/skills/payments_agent/stripe-skill. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Stripe Skill?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Stripe Skill?

As many as you like. As long as a model supports skills, you can use Stripe Skill with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Stripe Skill AI skill free?

Yes. It is published on GitHub by zeenie-ai under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇