Tech Stack Detector: Website Technology Lookup

Find out what any website is built with. Give Tech Stack Detector a list of websites and get one row per site with its CMS, shop platform, analytics, CDN, hosting, email provider and 190+ other technologies, each with the evidence that matched and a confidence score. Sites that fail to load are listed separately and are never charged.

What is Tech Stack Detector?

Tech Stack Detector is a website technology lookup tool, an alternative to BuiltWith or Wappalyzer for lists of sites you want to check with the API or on a schedule. It reads each site's HTML, HTTP headers, cookies, script and stylesheet URLs, meta tags, robots.txt and DNS records (CNAME, NS, MX), and runs a real browser only when a page needs JavaScript to show its content.

• One row per website (or one row per website and technology, for spreadsheets).
• Main columns ready to filter on: cms, ecommercePlatform, analytics, cdn, hosting, emailProvider.
• Evidence for every detection: the header, cookie, script URL, meta tag, DNS record or JS global that matched.
• Up to 10,000 websites per run. HTTP first, so 1,000 sites finish in minutes at 1 GB of memory.

What can it detect?

195 technologies in 34 categories:

Category	Technologies
Email Provider (18)	Amazon SES, Barracuda, Cloudflare Email Routing, Fastmail, GoDaddy Email, Google Workspace, iCloud Mail, Mailgun, Microsoft 365, Mimecast, Namecheap Private Email, Proofpoint, Proton Mail, Rackspace Email, Titan Email, Yahoo Mail, Yandex 360, Zoho Mail
Analytics (13)	Amplitude, Clarity, Fathom, FullStory, Google Analytics, Google Analytics 4, Heap, Hotjar, Matomo, Mixpanel, Plausible, PostHog, Segment
CMS (13)	Contentful, Drupal, Ghost, HubSpot CMS, Joomla, Magento, PrestaShop, Shopify, Squarespace, Strapi, Webflow, Wix, WordPress
JavaScript Framework (13)	Angular, Astro, Ember.js, Gatsby, Next.js, Nuxt.js, Preact, React, Remix, Stimulus, Svelte, SvelteKit, Vue.js
Hosting (12)	AWS, Azure, DigitalOcean, Fly.io, GitHub Pages, Google Cloud, Heroku, Netlify, Railway, Render, Vercel, WP Engine
JavaScript Library (10)	Alpine.js, Axios, Backbone.js, Chart.js, D3.js, HTMX, jQuery, Lodash, Moment.js, Turbo
Programming Language (9)	ASP.NET, Django, Express.js, Flask, Laravel, PHP, Ruby on Rails, Spring, TypeScript
CDN (8)	Akamai, AWS CloudFront, Cloudflare, Fastly, jsDelivr, KeyCDN, StackPath, unpkg
CSS Framework (8)	Ant Design, Bootstrap, Bulma, Chakra UI, Foundation, Material UI, Semantic UI, Tailwind CSS
Chat & Support (8)	Crisp, Drift, Freshdesk, Intercom, LiveChat, Tawk.to, Tidio, Zendesk
Advertising (7)	Criteo, Facebook Pixel, Google Ads, LinkedIn Insight Tag, Pinterest Tag, TikTok Pixel, Twitter Pixel
Email Marketing (6)	ActiveCampaign, ConvertKit, HubSpot, Klaviyo, Mailchimp, SendGrid
Payment (6)	Afterpay, Braintree, Klarna, PayPal, Square, Stripe
Security (6)	Cloudflare Turnstile, hCaptcha, reCAPTCHA, Recaptcha Enterprise, Sucuri, Wordfence
A/B Testing (5)	AB Tasty, Google Optimize, LaunchDarkly, Optimizely, VWO
Animation (4)	Framer Motion, GSAP, Lottie, Three.js
Build Tool (4)	Parcel, Turbopack, Vite, Webpack
Error Tracking (4)	Bugsnag, Datadog RUM, LogRocket, Sentry
SEO (4)	Open Graph, Schema.org, Twitter Cards, Yoast SEO
Tag Manager (4)	Adobe Launch, Cloudflare Zaraz, Google Tag Manager, Tealium
Web Server (4)	Apache, LiteSpeed, Nginx, Varnish
Cookie Consent (3)	Cookie Notice, Cookiebot, OneTrust
E-commerce (3)	BigCommerce, OpenCart, Woocommerce
Font/Icon Library (3)	Font Awesome, Google Fonts, Typekit/Adobe Fonts
Maps (3)	Google Maps, Leaflet, Mapbox
Page Builder (3)	Beaver Builder, Divi, Elementor
API (2)	Apollo Client, GraphQL
Authentication (2)	Auth0, Okta
Backend (2)	Firebase, Supabase
PWA (2)	PWA, Service Worker
Performance (2)	AMP, Lazy Loading
UI Component (2)	Slick Slider, Swiper
Edge Computing (1)	Cloudflare Workers
Real-time (1)	Socket.io

How to find what a website is built with

1. Open the Actor and paste your websites into Websites, one per line. Bare domains such as wordpress.org work.
2. Keep Mode on auto unless you need every site rendered in a browser.
3. Click Start. Each site takes 1 to 3 seconds in the HTTP pass.
4. Open the Output tab, filter on cms, ecommercePlatform or any other column, and export to CSV, Excel or JSON, or read the results with the Apify API.

How much does it cost to check 1,000 websites?

You pay per website analysed. Websites that fail to load (DNS errors, timeouts, error pages) are free and listed in the FAILED_URLS record of the run's key-value store. There is no extra start fee beyond the standard Apify run start.

Event	Charged when	Free plan	Bronze	Silver	Gold
Site analysed (site-analysed)	One website analysed and its result saved. Sites that fail to load are not charged.	2.00 USD per 1,000 events	2.00 USD per 1,000 events	1.80 USD per 1,000 events	1.60 USD per 1,000 events
Actor start (apify-actor-start)	Charged by Apify when the run starts: once up to 1 GB of memory, then once for each extra GB.	0.05 USD per 1,000 runs	0.05 USD per 1,000 runs	0.05 USD per 1,000 runs	0.05 USD per 1,000 runs

Example: 1,000 websites on the Bronze plan cost 1,000 x 0.002 USD = 2.00 USD, plus 0.00005 USD for the run start. Platform usage (compute, storage, traffic) is included in these prices.

Free plan: the 5 USD of monthly Apify credit covers about 2,500 websites a month when no other Actor is used.

You can set a maximum cost per run in the run options. The Actor stops starting new sites when the limit is reached, finishes the sites in progress and writes its summary.

Input example

{
    "urls": ["https://www.shopify.com", "wordpress.org", "https://react.dev"],
    "mode": "auto",
    "outputFormat": "site",
    "minConfidence": 30,
    "includeEvidence": true,
    "checkDns": true,
    "checkRobotsTxt": true,
    "maxConcurrency": 10
}

Output example: site row

A row from a real run (the technologies list is shortened here):

{
    "url": "https://wordpress.org",
    "finalUrl": "https://wordpress.org",
    "domain": "wordpress.org",
    "statusCode": 200,
    "https": true,
    "technologiesCount": 6,
    "categories": "CMS, SEO, Tag Manager, Web Server",
    "technologies": [
        {
            "name": "WordPress",
            "category": "CMS",
            "version": "7.2",
            "confidence": 100,
            "website": "https://wordpress.org",
            "evidence": [
                { "type": "html", "match": "wp-content/", "matches": ["wp-content/"] },
                { "type": "meta", "match": "generator=WordPress 7.2-beta5-65438", "matches": ["generator=WordPress 7.2-beta5-65438"] }
            ]
        },
        {
            "name": "Nginx",
            "category": "Web Server",
            "version": null,
            "confidence": 60,
            "website": "https://nginx.org",
            "evidence": [{ "type": "headers", "match": "server: nginx", "matches": ["server: nginx"] }]
        }
    ],
    "technologyGroups": {
        "SEO": ["Open Graph", "Twitter Cards", "Schema.org"],
        "CMS": ["WordPress"],
        "Web Server": ["Nginx"],
        "Tag Manager": ["Google Tag Manager"]
    },
    "cms": "WordPress",
    "ecommercePlatform": null,
    "analytics": [],
    "cdn": null,
    "hosting": null,
    "emailProvider": null,
    "renderedWithBrowser": false,
    "redirectedToOtherDomain": false,
    "blockedByChallenge": false,
    "analysisTimeMs": 2080,
    "analyzedAt": "2026-11-12T21:12:56.492Z"
}

Output example: technology row

With "outputFormat": "technology" each technology gets its own row. The price stays one event per website.

{
    "url": "https://www.shopify.com",
    "finalUrl": "https://www.shopify.com",
    "domain": "www.shopify.com",
    "statusCode": 200,
    "https": true,
    "cms": "Shopify",
    "ecommercePlatform": "Shopify",
    "cdn": "Cloudflare",
    "hosting": null,
    "emailProvider": "Google Workspace",
    "renderedWithBrowser": false,
    "redirectedToOtherDomain": false,
    "blockedByChallenge": false,
    "analyzedAt": "2026-11-12T21:12:59.123Z",
    "technology": "Shopify",
    "category": "CMS",
    "version": null,
    "confidence": 85,
    "website": "https://shopify.com",
    "evidence": "html: cdn.shopify.com | scripts: https://cdn.shopify.com/shopifycloud/world/brochure/production-shopify/assets/entry.client-DwpyJ0Wk.js"
}

Output fields

Field	Type	Description
url	string	Input URL, normalised (https:// added, host in lower case).
finalUrl	string	URL of the analysed page after redirects.
domain	string	Host name of the input URL.
statusCode	integer or null	HTTP status code of the final page.
https	boolean	True when the final URL uses HTTPS.
technologies	array	Site rows only. Each detected technology with its category, version (or null), confidence from 0 to 100, evidence and website.
technologiesCount	integer	Site rows only. Number of technologies at or above the minimum confidence.
categories	string	Site rows only. Comma-separated list of the categories found.
technologyGroups	object	Site rows only. Category name to list of technology names.
technology	string or null	Technology rows only. Name of the technology (null when the site shows none).
category	string or null	Technology rows only. Category of the technology.
version	string or null	Technology rows only. Version when the site exposes it, else null.
confidence	integer or null	Technology rows only. Confidence from 0 to 100.
website	string or null	Technology rows only. Home page of the technology.
evidence	string	Technology rows only. Matched signals as text, for example scripts: https://cdn.shopify.com/...
cms	string or null	Main content management system, or null.
ecommercePlatform	string or null	Main e-commerce platform, or null.
analytics	array	Site rows only. Analytics tools found.
cdn	string or null	Main CDN, or null.
hosting	string or null	Hosting or platform-as-a-service provider, or null.
emailProvider	string or null	Email provider from the domain's MX records, or null.
renderedWithBrowser	boolean	True when the browser pass ran for this site.
redirectedToOtherDomain	boolean	True when the input URL redirects to another site (for example a parked domain).
blockedByChallenge	boolean	True when the site showed a bot challenge page; the row then holds only what that page revealed (often the CDN).
analysisTimeMs	integer	Site rows only. Time spent on the site in milliseconds.
analyzedAt	string	ISO 8601 time of the analysis.

The run also writes two records to its key-value store: SUMMARY (counts per technology and category, sites analysed, failed and not processed) and FAILED_URLS (each failed site with the reason and the number of attempts).

Modes: auto, http, browser

• auto (default): the HTTP pass reads every site. A site goes to the browser pass only when its page is empty, is a small client-rendered shell (for example a single <div id="root"> with script bundles), shows fewer than 2 technologies, returns 401, 403, 429 or 5xx, or shows a bot challenge.
• http: HTTP pass only. Fastest and cheapest for the platform; JavaScript globals are not checked.
• browser: every site is rendered in Chromium (images, fonts and media are blocked). Use 2,048 MB of memory or more; at 1,024 MB the Actor renders 1 page at a time, at 2,048 MB 3 pages, at 4,096 MB 6 pages.

Use cases

• Lead lists by platform: keep only the Shopify or WooCommerce stores in a list of prospects, or every site on HubSpot.
• Competitor research: see which analytics, A/B testing and chat tools competitors run.
• Audits: check which CDN, hosting and email provider a list of client sites uses, and spot old jQuery or CMS versions.
• Enrichment pipelines: call the Actor with the API or from Make or n8n for each new lead, or schedule it weekly on a saved list.

Limits

• Detection is signature-based: technologies that leave no trace in HTML, headers, cookies, DNS or JS globals are missed.
• Sites behind bot challenges return few results (often only the CDN); no bypass is attempted. Such rows have blockedByChallenge: true.
• Versions are shown only when the site exposes them.
• Only the page you list is read (plus robots.txt and DNS); other pages of the site are not crawled.
• At most 3 MB of HTML is read per page.
• URLs that point to private or local network addresses are refused.

FAQ

Am I charged for sites that fail? No. Only websites whose result row is saved are charged. A site that loads but shows no technology above your minimum confidence is saved with technologies: [] and is charged, because the analysis was delivered.

What does the confidence score mean? Each matching signal adds a weight (for example a meta generator tag or a vendor header weighs more than a word in the HTML). The score is the sum, capped at 100. Raise Minimum confidence to keep only strong matches.

Why is a technology marked "implied"? Some technologies imply others: a Next.js site runs React, a WooCommerce shop runs WordPress. The implied one gets the confidence of the one that implies it.

I used version 1. Will my integration break? No field was removed or renamed. See the changelog for the differences.

Other Actors by the same author

• App Store and Google Play Scraper: app details and reviews from both stores.
• All Actors: apify.com/magicfingers