A TXL STUDIO PRODUCT

Know what's actually wrong before someone else finds it.

TXL Aegis scans your live site for real, fixable security gaps — missing headers, weak rate limits, exposed files — and hands you a plain-English grade and a checklist, not a wall of jargon.

Authorized, non-destructive scanning only — every scan requires confirmed permission first.
txl-aegis

Fifteen areas, dozens of checks — all non-destructive, all with a real fix attached.

Most scanners tell you a header is missing. Aegis tests whether your protections actually work — is the CSP strong, is the CDN script pinned, is that old subdomain still pointing at something you don’t own. No exploitation, no credential attacks: every finding ships with the exact config line to change, tagged against OWASP Top 10:2025 and CWE.

Web hardening

CSP quality

Not just “is there a CSP” — flags unsafe-inline, wildcard sources, missing object-src / base-uri, and report-only policies that block nothing.

Web hardening

HSTS & HTTPS upgrade

Checks HSTS strength (max-age, subdomains, preload eligibility) and that plain HTTP is actually redirected to HTTPS.

Web hardening

CORS misconfiguration

Sends a probe from an untrusted origin to see whether your server reflects it — the dangerous case is with credentials allowed.

Web hardening

Headers, cookies & methods

Security headers, Secure / HttpOnly / SameSite cookies, risky HTTP verbs and mixed content — each with the exact fix.

Supply chain

Third-party scripts & SRI

Inventories every outside host your page runs code from, and flags pinnable scripts that are missing Subresource Integrity.

Supply chain

Compromised CDNs

Flags scripts loaded from domains tied to publicly reported supply-chain attacks, like the 2024 polyfill.io incident.

Supply chain

Outdated JavaScript libraries

Spots end-of-life or known-vulnerable jQuery, Bootstrap, AngularJS, Lodash and more — with the CVE and the fix.

Email & domain

Email spoofing

SPF, DKIM and DMARC — whether anyone could send mail that looks like it came from your domain, including policy strength.

Email & domain

DNS & mail hardening

CAA, DNSSEC, MTA-STS and TLS-RPT — the settings most basic scanners never look at.

Email & domain

Domain expiry watch

Warns before your registration lapses — the quiet failure that takes the website and email down together.

Attack surface

Subdomain discovery

Finds subdomains in public certificate-transparency logs and flags internal-looking names that are exposed to the world.

Attack surface

Dangling DNS & takeover risk

Detects DNS records that point at services which no longer exist — the setup behind subdomain takeovers.

Attack surface

Exposed files & open ports

.env, .git and backups; risky services like databases or RDP; and whether a security.txt tells researchers how to reach you.

Transport & auth

TLS & legacy protocols

Certificate expiry and chain, plus whether the server still accepts deprecated TLS 1.0 / 1.1.

Transport & auth

Rate limiting & brute-force defense

A bounded probe with throwaway credentials checks whether login lockout or throttling exists at all — it never guesses a real password.

Three steps, and you're not waiting on a consultant's calendar.

01

Scan

I confirm permission first, then run the checks against your live site — a few minutes, nothing invasive.

02

Grade & fix list

An A–F grade plus a prioritized "what to fix" list — each item with the actual config line, not a vague warning. Exportable as a PDF you can hand to a client or a dev.

03

Re-verify, or stay on it

Ship the fixes, I re-scan, you get proof it's resolved. Or set it to check itself on a schedule, with an alert if anything regresses.

This is the actual output — grade, priorities, real fixes.

Example data below, not a real client's — this is exactly the format every scan produces.

example-shop.com
Security scan report
B
84 / 100
0 high · 1 medium · 2 low
MEDIUM
No rate limiting observed
Add rate limiting at your edge/proxy or in application middleware, scoped to login, checkout, and contact endpoints.
LOW
Missing header: Content-Security-Policy
Add a CSP header starting from default-src 'self', with exceptions only for domains you actually load scripts/styles from.
LOW
Missing header: Permissions-Policy
Restrict unused browser features — camera, microphone, geolocation — that the site never actually uses.
generated_at: illustrative example · not a live scan result

I ship the sites. I got tired of shipping them unhardened.

TXL Studio builds websites, dashboards, and AI-powered chatbots for businesses. TXL Aegis grew out of a habit from that work: check what I ship, not just that it works. Now it's available as its own thing — for any site, not only the ones I built.

It's not a penetration-testing toolkit and it doesn't pretend to catch everything. It's the fast, honest first pass every live site should get.

Want to check by hand first? Here’s the 9-point website security checklist.

✓
Authorization required, always. Every scan needs confirmed permission before it runs — no exceptions.
✓
Non-destructive by design. Read-only checks and a handful of bounded probes — nothing that can break a live site.
✓
Plain-English output. A grade and a fix list a business owner can understand, not just a developer.
✓
Built alongside real client work. Shaped by shipping actual e-commerce and business sites, not written in the abstract.

Complimentary with a build. Standalone on its own.

No retainers, no recurring "security platform" fee — just a scan, a fix, and proof it's resolved.

With a new build

Included, not an upsell

Commission a website, dashboard, or chatbot and the Aegis scan, fixes, and verified report ship with it — no extra charge.

Complimentary
Standalone

TXL Aegis report

For any live site, built by me or not. I implement the fixes (where you give access), re-scan, and hand over a signed-off "Secured" report to keep or share.

₹4,000starting