Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Automate Stripe tasks via Rube MCP (Composio): customers, charges, subscriptions, invoices, products, refunds. Always search tools first for current schemas.
.claude/skills/davepoon-stripe-automation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | 253% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 114% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 551% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 12% | 0% |
| case-11 | ✓→✗ | ▼ Worse | 172% | 0% |
Automate Stripe payment operations through Composio's Stripe toolkit via Rube MCP.
Toolkit docs: composio.dev/toolkits/stripe
RUBE_MANAGE_CONNECTIONS with toolkit stripeRUBE_SEARCH_TOOLS first to get current tool schemasGet Rube MCP: Add https://rube.app/mcp as an MCP server in your client configuration. No API keys needed — just add the endpoint and it works.
RUBE_SEARCH_TOOLS respondsRUBE_MANAGE_CONNECTIONS with toolkit stripeWhen to use: User wants to create, update, search, or list Stripe customers
Tool sequence:
STRIPE_SEARCH_CUSTOMERS - Search customers by email/name Optional]STRIPE_LIST_CUSTOMERS - List all customers Optional]STRIPE_CREATE_CUSTOMER - Create a new customer Optional]STRIPE_POST_CUSTOMERS_CUSTOMER - Update a customer Optional]Key parameters:
email: Customer emailname: Customer namedescription: Customer descriptionmetadata: Key-value metadata pairscustomer: Customer ID for updates (e.g., 'cus_xxx')Pitfalls:
When to use: User wants to create charges, payment intents, or view charge history
Tool sequence:
STRIPE_LIST_CHARGES - List charges with filters Optional]STRIPE_CREATE_PAYMENT_INTENT - Create a payment intent Optional]STRIPE_CONFIRM_PAYMENT_INTENT - Confirm a payment intent Optional]STRIPE_POST_CHARGES - Create a direct charge Optional]STRIPE_CAPTURE_CHARGE - Capture an authorized charge Optional]Key parameters:
amount: Amount in smallest currency unit (e.g., cents for USD)currency: Three-letter ISO currency code (e.g., 'usd')customer: Customer IDpayment_method: Payment method IDdescription: Charge descriptionPitfalls:
When to use: User wants to create, list, update, or cancel subscriptions
Tool sequence:
STRIPE_LIST_SUBSCRIPTIONS - List subscriptions Optional]STRIPE_POST_CUSTOMERS_CUSTOMER_SUBSCRIPTIONS - Create subscription Optional]STRIPE_RETRIEVE_SUBSCRIPTION - Get subscription details Optional]STRIPE_UPDATE_SUBSCRIPTION - Modify subscription Optional]Key parameters:
customer: Customer IDitems: Array of price items (price_id and quantity)subscription: Subscription ID for retrieval/update (e.g., 'sub_xxx')Pitfalls:
When to use: User wants to create, list, or search invoices
Tool sequence:
STRIPE_LIST_INVOICES - List invoices Optional]STRIPE_SEARCH_INVOICES - Search invoices Optional]STRIPE_CREATE_INVOICE - Create an invoice Optional]Key parameters:
customer: Customer ID for invoicecollection_method: 'charge_automatically' or 'send_invoice'days_until_due: Days until invoice is duePitfalls:
auto_advance: false for draft invoicesWhen to use: User wants to list or search products and their pricing
Tool sequence:
STRIPE_LIST_PRODUCTS - List products Optional]STRIPE_SEARCH_PRODUCTS - Search products Optional]STRIPE_LIST_PRICES - List prices Optional]STRIPE_GET_PRICES_SEARCH - Search prices Optional]Key parameters:
active: Filter by active/inactive statusquery: Search query for search endpointsPitfalls:
When to use: User wants to issue refunds on charges
Tool sequence:
STRIPE_LIST_REFUNDS - List refunds Optional]STRIPE_POST_CHARGES_CHARGE_REFUNDS - Create a refund Optional]STRIPE_CREATE_REFUND - Create refund via payment intent Optional]Key parameters:
charge: Charge ID for refundamount: Partial refund amount (omit for full refund)reason: Refund reason ('duplicate', 'fraudulent', 'requested_by_customer')Pitfalls:
Stripe uses smallest currency unit:
limit parameter (max 100)has_more in responsestarting_after with last object ID for next pagehas_more is falseAmount Units:
ID Prefixes:
cus_, Charges: ch_, Subscriptions: sub_in_, Products: prod_, Prices: price_pi_, Refunds: re_| Task | Tool Slug | Key Params | |------|-----------|------------| | Create customer | STRIPE_CREATE_CUSTOMER | email, name | | Search customers | STRIPE_SEARCH_CUSTOMERS | query | | Update customer | STRIPE_POST_CUSTOMERS_CUSTOMER | customer, fields | | List charges | STRIPE_LIST_CHARGES | customer, limit | | Create payment intent | STRIPE_CREATE_PAYMENT_INTENT | amount, currency | | Confirm payment | STRIPE_CONFIRM_PAYMENT_INTENT | payment_intent | | List subscriptions | STRIPE_LIST_SUBSCRIPTIONS | customer | | Create subscription | STRIPE_POST_CUSTOMERS_CUSTOMER_SUBSCRIPTIONS | customer, items | | Update subscription | STRIPE_UPDATE_SUBSCRIPTION | subscription, fields | | List invoices | STRIPE_LIST_INVOICES | customer | | Create invoice | STRIPE_CREATE_INVOICE | customer | | Search invoices | STRIPE_SEARCH_INVOICES | query | | List products | STRIPE_LIST_PRODUCTS | active | | Search products | STRIPE_SEARCH_PRODUCTS | query | | List prices | STRIPE_LIST_PRICES | product | | Search prices | STRIPE_GET_PRICES_SEARCH | query | | List refunds | STRIPE_LIST_REFUNDS | charge | | Create refund | STRIPE_CREATE_REFUND | charge, amount | | Payment methods | STRIPE_LIST_CUSTOMER_PAYMENT_METHODS | customer | | Checkout session | STRIPE_CREATE_CHECKOUT_SESSION | line_items | | List payment intents | STRIPE_LIST_PAYMENT_INTENTS | customer |
Powered by Composio
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→fail | 2,335 | 7,641 | +227% | 1 | 1 | 0% | 388 | 2,524 | +551% | 0 | 0 | — |
case-18 | pass→pass | 9,824 | 7,057 | -28% | 1 | 1 | 0% | 1,999 | 3,406 | +70% | 0 | 0 | — |
case-19 | fail→fail | 6,882 | 5,073 | -26% | 1 | 1 | 0% | 1,289 | 2,241 | +74% | 0 | 0 | — |
case-01 | fail→fail | 6,176 | 6,077 | -2% | 1 | 1 | 0% | 457 | 2,381 | +421% | 0 | 0 | — |
case-20 | fail→pass | 3,492 | 1,723 | -51% | 1 | 1 | 0% | 643 | 2,269 | +253% | 0 | 0 | — |
case-02 | fail→fail | 9,112 | 5,561 | -39% | 1 | 1 | 0% | 1,754 | 2,413 | +38% | 0 | 0 | — |
case-03 | fail→fail | 5,270 | 14,828 | +181% | 1 | 1 | 0% | 999 | 2,474 | +148% | 0 | 0 | — |
case-04 | pass→pass | 11,626 | 8,167 | -30% | 1 | 1 | 0% | 2,290 | 3,558 | +55% | 0 | 0 | — |
case-05 | pass→fail | 11,309 | 6,880 | -39% | 1 | 1 | 0% | 2,241 | 2,508 | +12% | 0 | 0 | — |
case-06 | pass→pass | 15,848 | 9,235 | -42% | 1 | 1 | 0% | 2,182 | 3,794 | +74% | 0 | 0 | — |
case-07 | fail→pass | 7,763 | 5,097 | -34% | 1 | 1 | 0% | 1,384 | 2,960 | +114% | 0 | 0 | — |
case-08 | pass→pass | 2,672 | 2,951 | +10% | 1 | 1 | 0% | 486 | 2,574 | +430% | 0 | 0 | — |
case-09 | pass→pass | 5,083 | 3,130 | -38% | 1 | 1 | 0% | 767 | 2,538 | +231% | 0 | 0 | — |
case-10 | pass→pass | 5,249 | 3,671 | -30% | 1 | 1 | 0% | 946 | 2,768 | +193% | 0 | 0 | — |
case-11 | pass→fail | 5,062 | 5,657 | +12% | 1 | 1 | 0% | 856 | 2,328 | +172% | 0 | 0 | — |
case-12 | fail→fail | 6,152 | 5,381 | -13% | 1 | 1 | 0% | 1,103 | 2,324 | +111% | 0 | 0 | — |
case-13 | pass→pass | 4,343 | 3,127 | -28% | 1 | 1 | 0% | 642 | 2,541 | +296% | 0 | 0 | — |
case-21 | fail→fail | 6,595 | 5,024 | -24% | 1 | 1 | 0% | 1,013 | 2,316 | +129% | 0 | 0 | — |
case-14 | pass→pass | 4,722 | 2,073 | -56% | 1 | 1 | 0% | 689 | 2,313 | +236% | 0 | 0 | — |
case-15 | pass→pass | 6,105 | 6,915 | +13% | 1 | 1 | 0% | 970 | 2,766 | +185% | 0 | 0 | — |
case-16 | pass→pass | 2,400 | 2,136 | -11% | 1 | 1 | 0% | 390 | 2,335 | +499% | 0 | 0 | — |
case-17 | fail→fail | 4,388 | 1,694 | -61% | 1 | 1 | 0% | 885 | 2,295 | +159% | 0 | 0 | — |
case-23 | pass→pass | 2,696 | 2,457 | -9% | 1 | 1 | 0% | 365 | 2,428 | +565% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 15 counted toward the lift figure. The other 8 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -4 percentage points is the difference between those two pass rates over the 15 comparable cases. 5 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.