SpunkArt
Websites, APIs and search

How to tell what technology a website is built with

Most websites leave fingerprints: a header, a folder name, a cookie, a comment in the HTML. With a browser and a terminal you can read those clues yourself and make an informed guess about the tech stack.

Why people look

Developers check stacks to learn how a site they admire was built. Agencies check before quoting a redesign. Security teams check to know which updates matter. Whatever the reason, remember that every clue can be hidden or faked, so your answer is an educated guess, not proof.

Clue 1: response headers

curl -sI example.com

Look for Server, X-Powered-By, Via and any header that starts with a vendor name. A CDN often adds its own cache or request-ID headers. Many well-run sites remove X-Powered-By on purpose, so its absence tells you little.

Clue 2: the page source

Press Ctrl+U (Cmd+Option+U on a Mac) and search for these patterns:

  • A <meta name="generator"> tag, which some content management systems add automatically.
  • Asset paths. Folder names such as /wp-content/, /_next/ or /_nuxt/ are strong hints for particular platforms and frameworks.
  • A root element like <div id="root"> or <div id="app"> with almost no text, which suggests a client-rendered single-page app.
  • Comments left by build tools or themes.

Clue 3: DevTools

Open DevTools and check three panels:

  1. Network: file names of scripts and stylesheets, and which domains they load from.
  2. Application (or Storage): cookie names. Session cookie names often reveal the server language or platform.
  3. Console: some frameworks expose a global variable you can type to confirm them.

Clue 4: DNS

dig example.com and dig example.com NS show where the domain points and who runs its name servers. A CNAME to a hosting provider’s domain, or name servers belonging to a CDN, tells you about hosting even when the HTML gives nothing away. Checking the MX record shows which email service the domain uses.

Putting the clues together

You seeReasonable guessConfidence
Generator tag naming a CMS plus matching asset foldersThat CMSHigh
Framework-specific build folder in script pathsThat frameworkHigh
Only a Server header with a generic nameThe web server, not the appLow
CNAME to a hosting platform domainHosted on that platformMedium to high
Empty HTML shell, content appears after scriptsClient-rendered appMedium

Mistakes to avoid

  • Confusing the CDN with the host. A CDN can sit in front of almost any origin server.
  • Assuming one page represents the whole site. Blogs, shops and help centres often run on separate systems under the same domain.
  • Treating a version number in a header as current. It may be stale or deliberately wrong.

Questions

Is it legal to inspect a site this way?

Reading what a public site sends to every visitor is normal browsing. Probing for hidden paths or vulnerabilities without permission is a different matter; do not do that.

How do I hide my own stack?

Remove version headers and generator tags. That reduces noise, but keeping software up to date matters far more than hiding it.