Content
72%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable — complete, correct, copy-paste-ready code for every Stripe integration pattern — and lean in its prose. Its weaknesses are structural: no explicit setup sequence or testing/validation checkpoints for a multi-step integration, no progressive disclosure (all ~310 lines inline in SKILL.md), and unquarantined time-sensitive version pins.
Suggestions
Add an explicit numbered setup workflow with validation checkpoints — e.g. install → configure env vars → create checkout → test locally with `stripe listen --forward-to localhost:3000/api/webhook/stripe` → verify session fulfillment in the database — so the multi-step integration sequence and its checkpoints are explicit rather than implicit.
Move the Stripe Elements and Embedded Checkout sections into separate reference files (e.g. `references/elements.md`, `references/embedded-checkout.md`) with clearly signaled one-level-deep links from SKILL.md, keeping the main file a lean overview.
Quarantine time-sensitive specifics — the pinned "2026-02-25.clover" apiVersion, "Recommended for 2026" server-action framing, and "Next.js 15+" — into a dedicated patterns/deprecated section or strip the pinned apiVersion so the SDK default is used.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is almost entirely executable code with minimal prose and no padding explaining concepts Claude already knows, but it embeds time-sensitive specifics outside any deprecated section — "apiVersion: '2026-02-25.clover'", "Server Action (Recommended for 2026)", "Next.js 15+" — and shows three full parallel checkout implementations. These are minor, trimmable issues fitting anchor 4 rather than the lean-every-token-earns-its-place bar of anchor 5. | 4 / 5 |
Actionability | Every section provides complete, copy-paste-ready TypeScript: checkout via server action, API route, and client redirect; a signature-verified webhook handler with raw-body handling; billing portal session creation; embedded checkout; Elements; and an env-var table. The examples cover the common cases exactly as the anchor-5 example does. | 5 / 5 |
Workflow Clarity | Sections follow a sensible implicit order (install → env vars → client init → checkout → webhooks → subscriptions) and webhook signature verification is a real checkpoint, but there is no explicit end-to-end sequence and no testing or validation guidance at all (no `stripe listen` for local webhook testing, test cards, or verifying session fulfillment). The missing checkpoints are more than minor, matching anchor 3 rather than anchor 4. | 3 / 5 |
Progressive Disclosure | Headers are well-organized, but the skill is ~310 lines with everything inline and no reference files at all; the Stripe Elements and Embedded Checkout sections are exactly the anchor-3 pattern of 'content that should be separate is inline'. Anchor 4 does not apply because there are no references to be 'mostly clear', and the file is well beyond the size where a single well-organized file scores 5. | 3 / 5 |
Total | 15 / 20 Passed |