| name | Payments |
| description | Integrate payments with provider selection, checkout flows, subscription billing, and security best practices. |
Situation Detection
| Context | Load |
|---|
| Choosing Stripe vs Paddle vs LemonSqueezy | providers.md |
| Implementing checkout, webhooks, refunds | integration.md |
| Subscription billing, trials, upgrades | subscriptions.md |
| PCI compliance, fraud prevention | security.md |
Universal Rules
Never store card data. Use provider-hosted checkout or tokenization. PCI compliance burden explodes the moment raw card numbers touch your server.
Webhooks are truth. Client-side success callbacks lie. A payment succeeded only when your webhook confirms it. Design for webhook-first verification.
Test mode exists for a reason. Use test cards, simulate failures, verify webhook handling. Production surprises cost real money and real customers.
Pricing psychology: $9.99/mo feels cheaper than $120/year, but annual retention is 2-3x higher. Default to annual with monthly option, not the reverse.
Provider Quick Compare
| Need | Recommendation |
|---|
| US/global B2C | Stripe (best docs, widest coverage) |
| SaaS selling to EU (VAT headache) | Paddle, LemonSqueezy (merchant of record) |
| Simple product, no dev resources | Gumroad, Lemonsqueezy hosted |
| Marketplace with splits | Stripe Connect |
| High-risk or adult | Specialized processors (CCBill, Epoch) |
See providers.md for detailed comparison.
Integration Checklist
Before going live:
Red Flags
- Storing CVV anywhere, ever → Instant PCI violation
- Trusting client-side payment confirmation → Fraud vector
- No retry logic for failed webhooks → Lost transactions
- Hardcoding prices in frontend → Easy manipulation
- Missing
cancel_at_period_end handling → Angry customers
When to Load More
| Situation | Reference |
|---|
| Evaluating payment processors | providers.md |
| Building checkout, handling webhooks | integration.md |
| Recurring billing, metering, trials | subscriptions.md |
| Fraud, PCI, chargebacks | security.md |
