Flat vector illustration of a website window being inspected through a magnifying glass with a shield

You can answer is my website secure in about thirty minutes, without being a security engineer. Run five layered checks in order: HTTP security headers, TLS configuration and certificate, DNS and email authentication, a passive external scan, then an authorised active scan. The first four are read-only and safe to run right now; the fifth needs your permission because it sends real test traffic to your own site.

Each step below names the exact tool, what a good result looks like and what a bad result means. Work through them in order, because a finding at one layer often explains a symptom at another.

Why this matters for your website

Most small-site compromises are not targeted. They come from automated scanners that crawl the internet looking for an outdated plugin, an exposed .env file or a login page with no rate limiting. Your site does not need to be interesting to be attacked; it only needs to be reachable and out of date.

The cost lands on the business, not the server. Hijacked checkout pages leak card data, spam injected into your pages destroys search rankings, and a mail domain without DMARC lets anyone send invoices in your name. A repeatable security check turns that risk into a short maintenance task.

The five checks, in order

Do them top to bottom. Steps 1 to 4 are passive and cannot break anything.

Diagram of five ordered website security checks from passive headers test to authorised active scan
Run the passive checks first; each one narrows what the active scan has to confirm.

Step 1: Check your HTTP security headers

Headers tell the browser how to defend your visitors. Run our free security headers checker, or curl -I https://yourdomain.com from a terminal. Compare the result against the OWASP Secure Headers Project.

  • Good: Strict-Transport-Security with a long max-age, a Content-Security-Policy without wildcard script sources, X-Content-Type-Options: nosniff, a Referrer-Policy, frame protection via frame-ancestors or X-Frame-Options, and a Permissions-Policy.
  • Bad: no HSTS means a visitor on public Wi-Fi can be downgraded to plain HTTP; no CSP means an injected script runs with full access to your pages and session cookies.
  • Also check: session cookies carry Secure, HttpOnly and SameSite.

Step 2: Test TLS and your certificate

Use Qualys SSL Labs or testssl.sh against your hostname and your www variant.

  • Good: TLS 1.2 and 1.3 only, a complete certificate chain, a hostname that matches the certificate, renewal automated well before expiry.
  • Bad: TLS 1.0 or 1.1 still enabled, an expired or self-signed certificate, or a chain that validates in Chrome but fails in older clients and payment callbacks.
  • Watch for: mixed content, where an HTTPS page loads an image or script over HTTP.

Step 3: Verify DNS and email authentication

Attackers impersonate brands through unprotected mail domains. Run our free SPF/DMARC checker, or query records with dig TXT yourdomain.com. DMARC is defined in RFC 7489.

  • Good: exactly one SPF record listing your real senders, a published DKIM selector, and a DMARC policy of quarantine or reject with a rua reporting address.
  • Bad: no SPF, two SPF records (which invalidates both), or p=none left in place for years so nothing is ever enforced.
  • Bad: a dangling CNAME pointing at a decommissioned SaaS host, which invites subdomain takeover.

Step 4: Run a passive external scan

This shows your site the way an automated attacker first sees it: subdomains, exposed files, software versions. Start with the free website security check.

  • Good: only subdomains you recognise, no directory listing, no reachable .git, .env, backup.zip or phpinfo.php, current CMS and plugin versions.
  • Bad: a forgotten staging site with real data, or a plugin version matching a known advisory in WPScan or the NVD.
  • Next action: patch or remove anything you no longer use before moving to step 5.

Step 5: Run an authorised active scan

Passive checks find exposure; active scans find exploitable behaviour. Tools such as OWASP ZAP, nuclei and WPScan send crafted requests to probe for injection, cross-site scripting, broken access control and outdated components. Only run them against assets you own or have written permission to test.

  • Good: no high or critical findings, forms reject malformed input, admin paths require authentication.
  • Bad: any reflected parameter that executes script, an admin endpoint reachable without a session, or a service on an open port that maps to a published CVE.
  • Practical note: scan a staging copy first if your production site is fragile, and take a backup either way.

Warning signs your site is already compromised

Some problems are past tense. Treat any of these as an incident, not a to-do item.

  • Visitors are redirected to unfamiliar domains, often only on mobile or only from search results.
  • An administrator account exists that nobody on your team created.
  • Google Search Console reports a security issue or your host suspends the account.
  • New PHP files appear in wp-content/uploads or timestamps change on core files you never edited.
  • Outbound spam, a sudden traffic spike from one country, or unexplained CPU load.

If you see these, take the site offline or into maintenance mode, rotate every credential, and restore from a known-good backup rather than deleting files one by one. Our complete website security guide covers cleanup and hardening in order.

Decision tree showing what to do when a website shows signs of compromise
Any confirmed indicator moves you from tuning to incident response.

How to check your own site

Scantinel (scantinel.ai) is an automated website security assessment platform that runs OSINT, network, and OWASP Top 10 web vulnerability tests against ownership-verified assets and delivers actionable, business-owner-legible reports.

The free tools above cover steps 1, 3 and 4 with no account. The Essential scan bundles those passive checks into one report: DNS and email posture, TLS audit, security headers, cookie flags, exposed common files, and a CMS version and known-vulnerable plugin check.

For step 5, the Full scan verifies ownership first, then adds active subdomain resolution and origin-IP discovery, port scanning with CVE correlation, a headless-browser crawl, an active OWASP Top 10 scan, a WordPress, WooCommerce, Drupal or Magento deep scan, and supply-chain checks for exposed .git or .env files and subdomain takeover. Scans are non-destructive and are not a penetration test. You can run a full scan when you are ready to go beyond passive checks.

Key takeaways

  • Check in layers: headers, TLS, DNS and email, passive exposure, then authorised active testing.
  • Steps 1 to 4 are read-only, free and safe to repeat monthly.
  • Missing HSTS, a p=none DMARC record and an outdated plugin are the three findings we see most often on small sites.
  • Active scanning requires ownership or written permission, and a backup before you start.
  • Redirects, unknown admin users and new files mean cleanup, not tuning.

Frequently asked questions

How can I check if a website is safe for free?

Run a free headers check, an SSL Labs test and a DNS and DMARC lookup; together they cover the exposure that automated attackers look for first. For your own site, add a passive scan that lists subdomains, exposed files and CMS plugin versions. All of these are read-only and take minutes.

How do I know if my website has been hacked?

Look for unexpected redirects, administrator accounts you did not create, security warnings in Google Search Console and new or modified files in your uploads and core directories. Outbound spam and sudden server load are supporting evidence. If two or more appear together, treat it as a compromise and restore from a clean backup.

Does a green padlock mean a site is secure?

No. The padlock only confirms that traffic is encrypted and that the certificate matches the hostname. A phishing page or a site running a vulnerable plugin can display exactly the same padlock, which is why steps 1 and 3 through 5 matter.

Published February 2025 by the Scantinel Security Research Team. Security guidance changes; verify specifics against primary sources such as OWASP, the NVD and WPScan.