Tokomu
A single-vendor online store blueprint built on Maincore. It starts as a shoe store and can be re-themed to other store types through config and data only, with an AI shopping assistant planned as its key differentiator.
One codebase, any store type: swap shoes for batik via config only
Live DemoThe Problem
Small businesses need an online store, but rebuilding the same catalog, cart and checkout from scratch for every product type is slow and repetitive. Shoppers also have to filter product lists manually instead of simply asking for what they need, for example "running shoes under Rp500,000".
Our Solution
Tokomu is a reusable single-vendor store blueprint. Store identity, categories, hero content and assets live in one typed configuration, so the UI components never change when the store type does. Prices, stock and orders are controlled entirely by the backend. An AI shopping assistant is planned that can only read real catalog data through backend tools, so it can't invent products or prices.
Key Features
- Responsive storefront with hero, categories and featured products
- Reusable product cards with automatic image fallback
- Search by name, combinable with category filter
- Price range filter and sorting
Tech Stack
- Next.js
- TypeScript
- NestJS
- PostgreSQL
- Prisma
- Tailwind CSS
- Turborepo
- Playwright
- Jest
- Docker
Results
Currently in active development, following a phased roadmap: static storefront, database-backed catalog, core commerce, hardening, then the AI assistant. Each phase ends with a quality gate covering typecheck, lint, tests with 70%+ coverage, end-to-end tests and security scans. Performance and AI-quality figures will be published only after they are measured.
Want something like this?
Tell us about your project and we will reply within one business day.
Chat on WhatsApp
