One-time payment, external checkout
Payment runs on Gumroad and Amazon; the site holds no subscriptions or card data.
Digital Products Shop (Global) · Case Study
ToolGenX — Claude Code Skills Packs Sold Once, Owned Forever
An English digital-products shop selling Claude Code skills packs, SwiftUI templates and AI toolkits for SEO, marketing, iOS and founders, with payment handled by Gumroad.
Results
Measurable Outcomes
19
Products
100+
Claude Code skills
$9-$119
Price range
Tek seferlik / One-time
Payment model
The Brief
What was the goal?
Users tired of renting AI tools by subscription needed a shop that says "buy once, keep forever". The site also had to rank for "Claude Code skills" searches on Google.
The Approach
Which methodology pillars I applied
The homepage was aligned to the "claude code skills" and "pack/bundle" keyword cluster after keyword research. Payment happens on Gumroad and Amazon, not on site; the site carries catalogue, content and trust signals.
See the full process → 47-Point AI-Ready SEO Audit methodology.
Tech Stack
What it's built with
- Next.js
- TypeScript
- Supabase
- Gumroad
- Schema.org Product
- Vercel
The Build
Key decisions and tradeoffs
The catalogue holds 19 products and 100+ skills, priced from $9 to $119. Every product page links to its Gumroad checkout; there is no on-site payment.
A server component in the shared header read the session cookie, which made the whole site dynamic on every request. Moving the login indicator to the client brought 50+ routes back to static generation.
21 blog posts were rewritten and 5 new Claude Code skills posts added, wired into a funnel from free tool to guide to product.
Steps Applied
Methodology in practice
Homepage aligned to a keyword cluster
Title and first section were written for "claude code skills" and "pack/bundle" queries.
Static generation
The cookie-reading shared component moved to the client; pages are generated at build time again.
Honest Reflection
What I'd Do Differently
8 of the 9 Wikidata IDs we put in the schema turned out to be wrong; one pointed to a Korean association instead of Stripe. They had been written without checking. One release also never went live because the Vercel project was connected to a different repo. Today I wouldn't call the job done without verifying every Wikidata ID through the official API and checking the live content after each deploy.
Screenshots
Want this for your project?
Same 47-point methodology, $499 fixed price, 5-7 business days. Book the audit.