Website Tech Stack Detector

Find out what any website is built on — and, uniquely, which e-commerce apps it actually runs.

Give it a list of domains. Get back the platform, CMS, framework, analytics, payment providers, consent tool and hosting — plus 295 e-commerce app fingerprints that generic detectors don't carry.

Never charged twice for the same domain

Every domain you receive is remembered across all your runs. Send the same list again next month and you pay only for what is new.

• Already analyzed → skipped, free, and the log says how many.
• Site down, blocked or unreachable → free, with the reason in a separate FAILED-URLS dataset.
• Want fresh data anyway? Set chargeForPreviouslyAnalyzed to true. Nothing hidden, nothing automatic.

There are no hidden row limits.

Why this one is different

Most detectors tell you "nginx, jQuery, Google Analytics" and stop. That's fine for a generic tech census, but useless if you're targeting DTC brands and need to know whether a store runs Klaviyo or Omnisend, Yotpo or Judge.me, Recharge or Bold.

This actor carries 295 e-commerce app fingerprints alongside the usual stack detection. If you sell to Shopify merchants, that's the column you actually need.

What you get per site

Field	Example
ecommerce_platform	Shopify, WooCommerce, Magento, Shopware, PrestaShop, BigCommerce, Salesforce Commerce Cloud, JTL-Shop …
cms	WordPress, Webflow, Squarespace, Wix, Drupal, TYPO3, Contentful, Sanity, Storyblok …
frameworks	Next.js, Nuxt, React, Vue, Angular, Svelte, Astro, Remix, Tailwind, Bootstrap …
analytics	GA4, Google Tag Manager, Meta Pixel, TikTok, Pinterest, LinkedIn, Clarity, Hotjar, Matomo, Plausible, Segment …
marketing_tools	Klaviyo, Mailchimp, HubSpot, ActiveCampaign, Brevo, Intercom, Zendesk, Crisp …
payment_providers	Stripe, PayPal, Klarna, Adyen, Mollie, Braintree, Afterpay, Apple Pay, Google Pay …
consent_platform	Cookiebot, OneTrust, Usercentrics, CookieYes, Complianz, Borlabs, Didomi …
hosting	Cloudflare, Vercel, Netlify, CloudFront, Fastly, Akamai, nginx, Apache, IIS …
ecommerce_apps	295 fingerprints with app name, category and Shopify App Store link
title, description, language, generator, server	page metadata

Honest limitations — please read

Static HTML only. This actor reads the HTML and response headers a server delivers. Technologies injected later by JavaScript, or loaded only after consent is granted, will not be seen. If you need full JS-rendered detection, use a browser-based tool — it will cost more and run slower, and it will find things this one misses.

Detection favours precision over recall. Every fingerprint requires a technical marker — a script URL, an asset path, a header — never just a brand name in the page text. An article about Shopify will not be reported as running Shopify. The trade-off is that unusual or self-hosted setups may be missed.

One request per site. No crawling of subpages, no proxy rotation, no anti-bot circumvention. Sites that block plain HTTP requests are reported as failures, and cost you nothing.

Input

{
  "urls": ["snocks.com", "vercel.com", "wordpress.org"],
  "includeEcommerceApps": true,
  "includeMeta": true
}

Example output

{
  "url": "snocks.com",
  "ecommerce_platform": "Shopify",
  "cms": null,
  "frameworks": [],
  "analytics": ["Google Tag Manager", "TikTok Pixel", "Pinterest Tag"],
  "marketing_tools": ["Klaviyo"],
  "payment_providers": ["PayPal", "Apple Pay"],
  "hosting": ["Cloudflare"],
  "ecommerce_apps": [
    { "name": "Klaviyo", "category": "Marketing", "appStoreUrl": "https://apps.shopify.com/klaviyo-email-marketing" }
  ],
  "technology_count": 11
}

Choose your fields

Set outputFields to just what you need — for example ["ecommerce_platform", "ecommerce_apps"] — and your export contains nothing else.

Pricing

Pay per successfully analyzed domain. Not per technology found, not per row.

Credits

E-commerce app fingerprints derived from feracommerce/shopify_app_detector (Apache-2.0). All platform, framework, analytics, payment, consent and hosting fingerprints are original work.

Support

A technology it should have found, or one it reported wrongly? Open an issue on this actor — false positives are treated as bugs and fixed.