Configures Claude Code hooks and Codex hooks.json/notify. Use when adding PreToolUse guards, Stop hooks, managed hooks, format-on-save, preflight, audits, or worktree/budget hooks.
日本語の概要は準備中です。原文の説明を表示しています。
Designs production payment and billing systems. Use when implementing Stripe, Paddle, Adyen, subscriptions, tax, marketplaces, or mobile purchase flows.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Use this skill to design, implement, and debug production payment integrations: checkout flows, subscription lifecycle handling, billing portals, entitlement gating, webhooks, regional pricing, and payment-testing strategy.
| Need | Default | Deviation | Reference |
|---|---|---|---|
| Standard SaaS subscription checkout | Stripe Checkout or Payment Links | Use Payment Element plus Express Checkout Element when branded custom UI is a hard requirement | references/stripe-patterns.md |
| Merchant of record (managed tax/seller-of-record) | Paddle or LemonSqueezy | Evaluate Stripe Managed Payments after checking seller location, product tax codes, account eligibility and supported surfaces at its eligibility page | references/platform-comparison.md |
| High-volume or enterprise multi-method processing | Adyen | — | references/platform-comparison.md |
| UK or EU bank debit and local methods | GoCardless, Mollie, or open-banking providers | Repo-dependent: no bundle source differentiates the three by situation; keep whichever the existing integration already uses | references/uk-eu-payments-guide.md |
| Usage-based metering or token billing | Stripe Billing Meters or Lago | — | references/subscription-lifecycle.md |
| Complex contract, seat, or revenue-recognition billing | Chargebee or Recurly on top of the processor | — | references/subscription-lifecycle.md |
| Webhook reliability | verified signatures plus idempotent processing | — | references/webhook-reliability-patterns.md |
| Entitlement and feature gating | registry plus API enforcement plus UI paywall | — | references/feature-gating-patterns.md |
| Native iOS in-app purchase | StoreKit 2 with backend JWS verification | Add RevenueCat only when multi-platform entitlement sync materially justifies it | references/storekit2-native-patterns.md |
| Checkout and webhook testing | Stripe CLI plus E2E automation | — | references/testing-patterns.md |
| AI agents initiating purchases | Scoped, limited-use payment credential plus seller-side approval | Human confirmation for high-risk actions | references/agentic-commerce.md |
Use this skill when the primary work is:
Route elsewhere when the main task is:
| Need | Use Instead |
|---|---|
| general backend implementation outside payments | ../software-backend/SKILL.md |
| API contract design without billing concerns | ../dev-api-design/SKILL.md |
| pricing strategy and package design | startup-business-models |
| checkout conversion optimization | marketing-cro |
| application security review | ../software-security-appsec/SKILL.md |
| native app implementation around store billing | ../software-mobile/SKILL.md |
| wallets, crypto settlement or on-chain money movement | ../software-crypto-web3/SKILL.md |
| Stripe-specific product/API guidance or account tooling | Available Stripe-maintained skills/plugins; discover them through Stripe’s agent guidance, then follow the runtime’s installation and permission policy |
payment_method_types unless you intentionally restrict available methods.See the Quick Reference table above for platform defaults by situation, plus references/platform-comparison.md and references/uk-eu-payments-guide.md for the detailed trade-offs. Mobile subscription default: StoreKit 2 or Play Billing as the store-of-record; add RevenueCat only when cross-platform entitlement sync, experiments, or web-plus-store coordination materially justify it.
Storefront.current and re-read the applicable guideline and regional terms before shipping. See references/storekit2-native-patterns.md.If Stripe is the chosen processor, apply the general Defaults above plus these Stripe-specific defaults:
Use references/stripe-patterns.md for API-version notes, webhook mappings, and migration detail.
When an AI agent buys on a user's behalf, or a seller accepts agent-initiated orders, give the agent only a per-transaction credential bound to a seller, currency, maximum amount, and expiry. Enforce spend limits in code, have the seller approve or decline against the final total, make every step idempotent under agent retries, require human confirmation for high-risk actions such as recurring charges or new merchants, and keep the consent record as dispute evidence. See references/agentic-commerce.md, which includes Stripe shared-payment-token notes.
Reconciliation gate.
Define a provider-independent money record for expected amount, currency, fee, tax, settlement, refund, dispute, and current entitlement. Reconcile webhooks and provider balance or settlement reports against that record on a schedule, with an exception queue and an owner. A successful checkout redirect or webhook is event evidence; neither is proof that cash, accounting state, and access rights agree.
Common mistakes:
| Mistake | Consequence |
|---|---|
| Trusting client redirect as proof of purchase | Provisioning before the verified webhook, creating access without payment |
| Hardcoding prices or plan logic in code | Catalog drift and deploy-gated pricing changes |
Setting payment_method_types unnecessarily | Unintentionally restricts available methods |
| Silently failing webhooks instead of retrying | Invisible entitlement loss or dangling subscriptions |
| Treating Apple/Google policy as stable across releases | Store rejection or forced redesign |
See: references/testing-patterns.md, references/webhook-reliability-patterns.md, references/feature-gating-patterns.md, references/ops-runbook-checkout-errors.md
Core
Regional and operational
Mobile and in-app purchase
Agentic commerce
Testing and migrations
Thresholds, dates, and legislative status in this area change often. Look them up at the source each time; this section keeps only the engineering controls.
stripe:test-cards at the time you write the test; do not copy them from memory.off_session: true plus the required consent/mandate captured during on-session setup, with authentication where required. A missing flag can cause unexpected renewal declines. Check EMVCo and provider support for published versus draft specification features before using them.When prior decisions or pitfalls are relevant, consult learnings.consolidated.md if present; use learnings.md only for needed history or as the available fallback. Otherwise skip both.
After applying it, if you encountered a pattern worth remembering, a mistake worth preventing, or a domain fact that surprised you, append one dated bullet to learnings.md via agents-skills-feedback-loop/scripts/append_learning.py. Do not modify SKILL.md itself.
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Configures Claude Code hooks and Codex hooks.json/notify. Use when adding PreToolUse guards, Stop hooks, managed hooks, format-on-save, preflight, audits, or worktree/budget hooks.
日本語の概要は準備中です。原文の説明を表示しています。
Configures and hardens Claude Code and Codex MCP servers. Use when connecting databases, APIs, SaaS, building servers, or serving a clearance-filtered knowledge base.
日本語の概要は準備中です。原文の説明を表示しています。
Owns instruction files: AGENTS.md, CLAUDE.md, personal and repo rules. Use when writing, pruning, auditing them, sharing rules across Claude and Codex, or fixing ignored rules.
日本語の概要は準備中です。原文の説明を表示しています。
Creates and audits agent skills: SKILL.md, references, scripts, runtime metadata. Use when writing, validating, or security-reviewing a skill, or fixing truncated skill listings.
日本語の概要は準備中です。原文の説明を表示しています。
Adds per-skill learnings loops for dated patterns, mistakes, and domain facts. Use when wiring skill memory, consolidation, or drift audits.
日本語の概要は準備中です。原文の説明を表示しています。
Chooses subagent, team, workflow, or debate and launches it on Claude Code or Codex. Use when delegating, running agent review boards, or installing shared agents.
日本語の概要は準備中です。原文の説明を表示しています。