Can I vibe code Shoplift?

Build an A/B testing engine specifically designed for Shopify stores using a Next.js App Router admin backend, a Cloudflare Worker for traffic distribution, and a lightweight vanilla JS client snippet. 1. DATA MODEL: Create PostgreSQL tables for `experiments` (id, name, status, target_url, allocation_pct), `variants` (id, experiment_id, name, template_handle, liquid_section_id), `impressions` (id, experiment_id, variant_id, visitor_id, timestamp), and `conversions` (id, experiment_id, variant_id, visitor_id, order_id, revenue, timestamp). 2. TRAFFIC SPLITTING EDGE LOGIC: Write a Cloudflare Worker that intercepts GET requests to target product/collection URLs. Check for an existing `ab_variant` cookie. If missing, deterministically assign a variant based on configured weights, set a 30-day cookie, and either rewrite the fetch request to the variant's Shopify alternate template handle (e.g., `/products/shoe-variant-b`) or add custom headers to prevent layout shift (FOUC). 3. EVENT TRACKING & ANALYTICS: Write a 1.5KB frontend JS SDK embedded via Shopify Theme App Extension that fires background `fetch` requests to `/api/v1/event` on variant render and add-to-cart clicks. Implement a Shopify Webhook handler for `orders/create` that reads the assigned variant cookie from the checkout attributes and attributes order total revenue to the variant. 4. STATISTICAL ENGINE: Build a Next.js dashboard view computing Bayesian probability to beat control, standard error, and projected revenue lift per visitor. 5. OUT OF SCOPE: Do not build a point-and-click WYSIWYG DOM visual builder; all variants must be managed by pointing to existing Shopify template handles or Liquid section IDs.

$ each button prefixes agent-specific run instructions · build your own product, never copy proprietary code, trademarks or designs