Give it a website. It fetches the homepage and identifies what's running it: CMS, ecommerce platform, JS framework, analytics/tag managers, CDN/hosting provider, payment widgets, and live chat widgets — from response headers, meta tags, and script signatures. No headless browser, no proxy.
Built for sales/competitive research (what platform is this prospect on?), agencies doing tech audits, and market research (who's using Shopify vs. WooCommerce in a given list of sites).
| Field | Type | Description |
|---|---|---|
startUrls |
array of URLs | Websites to analyze (homepage only, not a crawl). |
{
"startUrls": [{ "url": "https://example.com" }]
}One record per URL:
{
"url": "https://example.com",
"finalUrl": "https://example.com/",
"cms": ["WordPress"],
"ecommerce": ["WooCommerce"],
"jsFrameworks": ["jQuery"],
"analytics": ["Google Analytics", "Google Tag Manager"],
"cdnHosting": ["Cloudflare"],
"payment": ["Stripe"],
"liveChat": ["Intercom"],
"server": "cloudflare",
"poweredBy": "PHP/8.2",
"generator": "WordPress 6.7"
}Plain HTTP fetch via Crawlee's CheerioCrawler.
Detection is signature-based (src/signatures.js): checks the generator
meta tag, <script src> domains, response headers (Server,
X-Powered-By, CF-Ray, X-Vercel-Id, etc.), and a few HTML body markers
per technology. Same approach tools like Wappalyzer use, with a smaller,
hand-written signature set (~30 technologies across 7 categories) rather
than a comprehensive database — it'll correctly return "nothing detected"
for a signature it doesn't have rather than guessing.
Only reads what the site already serves publicly to any visitor's browser — no login, no evasion of any protection.
Add an entry to the SIGNATURES array in src/signatures.js: a name, a
category (one of cms, ecommerce, jsFrameworks, analytics,
cdnHosting, payment, liveChat), and a test(context) function where
context has headers (lowercased keys/values), html (lowercased full
source), scriptSrcs (lowercased <script src> values), and generator
(lowercased meta generator content, or '').