Google has Rich Results Test. This is the AI equivalent.
Scan any product page URL and find out if ChatGPT, Perplexity, and Gemini can discover, understand, and recommend your product — in 60 seconds.
Live Demo · Blog · Shopify App · 繁體中文
AI search engines are sending 4-5x higher converting traffic to e-commerce stores (Adobe, 2026). But most stores are invisible to them.
Traditional SEO won't help — ChatGPT doesn't look at your meta keywords or backlinks. It looks at structured data, crawler access, and content readability. Most Shopify stores have gaps in all three.
AIViz scans your product page and tells you exactly what's missing.
| Feature | Detail | |
|---|---|---|
| 📊 | 3-Layer Score (0-130 pts) | Crawler access · Structured data · Advanced optimization |
| 🔍 | AI Search Simulation | Simulates real queries — would ChatGPT recommend your product? |
| 🤖 | AI Readability Assessment | Gemini evaluates your page from an AI engine's perspective |
| 🔧 | Copy-Paste Fix Code | Get JSON-LD snippets you can add directly to your page |
| 🔗 | Shareable Reports | Each report gets a short URL (/r/:id), stored 30 days |
| 🌐 | Bilingual | Auto-detects English / 繁體中文 |
Running a Shopify store? SEO Checkup is our free Shopify App that scans all your products at once — no need to check URLs one by one.
| Rule | Pts | Checks |
|---|---|---|
| robots.txt | 10 | AI crawlers (GPTBot, PerplexityBot, etc.) not blocked |
| Meta Description | 5 | Present and adequate length |
| Image Alt Text | 5 | Descriptive alt attributes |
| JS Rendering | 5 | Content in raw HTML, not JS-only |
| Canonical URL | 5 | Correct canonical tag |
| Rule | Pts | Checks |
|---|---|---|
| Product Schema | 20 | schema.org/Product JSON-LD exists |
| Name & Description | 15 | Complete product info |
| Price & Currency | 15 | Valid pricing with currency |
| Product Image | 10 | Absolute URL, accessible |
| Aggregate Rating | 10 | Review/rating data |
| Brand Info | 10 | Brand name identifiable |
| Rule | Pts | Checks |
|---|---|---|
| Page Speed | 10 | Server response < 3s |
| llms.txt | 10 | llms.txt file present |
git clone https://github.com/Pin-Han/aiviz.git
cd aiviz
npm install
# Set up API env
cp api/.env.example api/.env
# Add your GEMINI_API_KEY (free tier: https://aistudio.google.com/apikey)
# Run frontend
cd frontend && npm run dev
# Run tests (74 tests)
npm testvercel link
vercel env add GEMINI_API_KEY
vercel --prodFor report sharing, add Upstash Redis from the Vercel Marketplace (prefix: KV).
aiviz/
├── frontend/ # React 19 SPA (Vite + Tailwind CSS)
│ ├── src/
│ │ ├── components/ # Report, ScoreCard, RadarChart, etc.
│ │ ├── blog/ # Markdown blog with i18n
│ │ ├── hooks/ # useAnalysis state machine
│ │ └── i18n/ # zh-TW / en translations
│ └── index.html
├── api/ # Vercel Serverless Functions
│ ├── analyze.ts # POST /api/analyze
│ ├── reports.ts # POST /api/reports
│ ├── reports/[id].ts # GET /api/reports/:id
│ ├── _lib/ # Crawler, parser, scorer, Gemini provider
│ └── _rules/ # Pluggable scoring rules
├── shared/ # TypeScript types & constants
└── vercel.json
Tech stack: React 19 · Vite · Tailwind CSS · Vercel Functions · Cheerio · Gemini 2.5 Flash · Upstash Redis · GA4
The scoring engine is pluggable:
// api/_rules/basic/my-rule.ts
import type { Rule } from '../../../shared/types.js'
export const myRule: Rule = {
id: 'my-rule',
name: 'My Rule',
description: 'Checks something useful',
category: 'basic',
maxScore: 10,
check(pageData) {
// Your logic here
return { score: 10, status: 'pass', message: 'All good' }
},
}Register it in api/_rules/index.ts and add tests in __tests__/.
Rule ideas: OG image dimensions · hreflang tags · FAQ schema · Breadcrumb schema · Review richness
- Web tool — scan any URL, get AI visibility report
- Shopify App — SEO Checkup
- Blog — AI visibility insights (/blog)
- AI Brand Monitoring — track brand mentions across AI search
- Competitor Analysis — compare AI visibility vs competitors
- Scheduled Scans — periodic checks with change alerts
MIT