Owner tools

For the owner of SafeExam (safexam.in).

Ownership

This launch is unclaimed.

Claim this launch to get an owner link, re-scan after fixes, track your score improvements, or remove your site from StackScope entirely.

Claim this launch

Fixes that improve your score

Security

HIGHAdd the missing security response headers
Missing 5 of 6 standard browser headers. Start with Referrer-Policy, X-Content-Type-Options, and X-Frame-Options (one line each), then roll out CSP in Report-Only mode.
WhyEach header limits a class of browser-side attack: clickjacking, XSS, MIME sniffing, plaintext fallback. Missing headers leave default-permissive behaviour in place.
WhereMost are one line each in your server config, reverse proxy, CDN, or framework headers.
Learn more
Why this matters & sources
Per-header consequence:
- Content-Security-Policy: Controls which scripts, styles, and fonts are allowed to run on your pages. Without it, an injected <script> tag runs with full access to your users' sessions and cookies.
- X-Frame-Options: Stops other sites loading yours inside an invisible iframe to trick users into clicking buttons they can't see (clickjacking).
- X-Content-Type-Options: Tells browsers not to guess file types. Without it, a user-uploaded image can be sniffed as JavaScript and executed.
- Referrer-Policy: Controls how much of the current URL is sent to third-party sites and analytics. The default leaks the full path, including query strings that may contain tokens or user IDs.
- Permissions-Policy: Turns off browser features your site doesn't use (camera, microphone, geolocation, payment) so an injected script can't trigger them.
CSP rollout: Content-Security-Policy is the hardest to get right. Start with Content-Security-Policy-Report-Only, which logs violations without blocking. Open DevTools Console, reload each page, and relax the policy until the console is clean. Rename the header to Content-Security-Policy to enforce. For real-user monitoring, point a report-to directive at a free collector (report-uri.com, Sentry, or a /api/csp-report route).

Optional improvements

These don't change your StackScope score but cover SEO, agent-readiness, security-researcher discoverability, and compliance items worth addressing.

Security

LOWAdd /.well-known/security.txt
WhyLets security researchers report vulnerabilities responsibly through a published contact channel.
WhereCreate /.well-known/security.txt with a Contact: email and Expires: date.
Learn more

Email security

MEDIUMMove DMARC from p=none to quarantine or reject
DMARC policy is p=none (monitoring only).
Whyp=none lets receivers log failures but still deliver spoofed mail. Real protection only kicks in at quarantine or reject.
WhereAfter reviewing a few weeks of rua= reports and confirming your real senders pass, change to p=quarantine (sends spoofs to spam) and then p=reject (blocks them outright).
Learn more
MEDIUMConfigure DKIM with your email provider
No DKIM record found at the common selectors we check.
WhyDKIM signs your outbound mail with a cryptographic key receivers can verify. Without it, receivers can't tell your real mail from a spoof.
WhereAsk your email provider for their DKIM setup. It's usually one TXT record at {selector}._domainkey.{your-domain}.
Learn more
Why this matters & sources
Selectors we probe: Resend, SendGrid, Postmark, Mailchimp, Brevo, Mailgun, HubSpot, Customer.io, ConvertKit, Cloudflare Email. If you use Google Workspace, Microsoft 365, or self-hosted email with a custom selector, you may already have DKIM published and we just don't see it.
LOWAdd MTA-STS
No MTA-STS DNS record published.
WhyPrevents mail to your domain being downgraded to plaintext mid-flight by a network attacker. Most launches don't have this, so deploying it puts you a tier above generic email-security checks.
WherePublish a TXT record at _mta-sts.{your-domain} plus a policy file at https://mta-sts.{your-domain}/.well-known/mta-sts.txt.
Learn more
LOWAdd a TLS-RPT record
No TLS-RPT record at _smtp._tls.{domain}.
WhyReceivers can tell you when STARTTLS handshakes to your mail server fail. Without it, silent TLS failures are invisible.
WherePublish one TXT record at _smtp._tls.{your-domain} like v=TLSRPTv1; rua=mailto:[email protected].
Learn more

Page basics & SEO

LOWTrim your <title> tag to under 60 characters
Currently 63 characters: "SafeExam | Online Exam Software for Colleges & Universities"
WhyGoogle truncates titles around 60 characters in search results, so anything past that gets cut off mid-sentence and costs click-through rate.
WhereAim for 50-60 characters that include your main keyword and brand.
Learn more
LOWTrim your meta description to under 160 characters
Currently 185 characters: "Affordable exam integrity for institutions: tab monitoring, violation tracking, question banks, and secure online examinations — more than Google Forms, less than enterprise proctoring."
WhyGoogle truncates around 155-160 characters on desktop SERPs (less on mobile), so anything past that won't appear in the snippet.
WhereEdit your <meta name="description"> tag. Put the most click-worthy phrase first.
Learn more

Agent / AI

LOWAdd an llms.txt file
WhyHelps AI models understand your site's content and how to use it. Not yet a standard but gaining adoption.
WhereCreate /llms.txt at the site root with a brief overview and key URLs.
Learn more
LOWAdd Link response headers
WhyLets agents discover your sitemap, privacy policy, and docs without parsing HTML, which most lightweight agents skip.
WhereSet Link: response headers in your server config or framework middleware.
Learn more

If a tip looks wrong (for example it says "add a consent banner" and you already have one) the detection's the bug, not you. StackScope sees what's public from the outside: HTTP response, rendered HTML, cookies, and DNS. We can miss vendors that load behind consent, are self-hosted, or use an install shape we haven't fingerprinted yet. Email [email protected] and we'll look into it.

Copy into Cursor, Claude, or ChatGPT

This prompt includes the detected stack and only the fixes StackScope found. It asks the AI to make concrete file-level changes, not a vague website review.

Score-affecting basics only. Ask your AI to handle these first; come back for the optional hardening once they're done.

Everything: score-affecting fixes plus optional email security, agent metadata, and best-practice items. Longer prompt, more for an "all in one" agent run.

I own https://safexam.in. StackScope (https://stackscope.dev) analysed it on 3 June 2026 and scored it 7.6/10 against their production-readiness checklist. Please help me apply the fixes below. These are the score-affecting basics. We'll circle back to the optional hardening (email security, advanced agent metadata) later.

Tech stack StackScope detected on the site (use this when suggesting where each fix lives in the codebase):
cloudflare, cloudflare-dns, cloudflare-email-routing, google-site-verification, hsts, http3, nextjs, react, tailwindcss, vercel

Score-affecting fixes (these improve the StackScope score; [HIGH] = most impact):
- [HIGH] Add the missing security response headers. Missing 5 of 6 standard browser headers. Start with `Referrer-Policy`, `X-Content-Type-Options`, and `X-Frame-Options` (one line each), then roll out CSP in `Report-Only` mode. Why: Each header limits a class of browser-side attack: clickjacking, XSS, MIME sniffing, plaintext fallback. Missing headers leave default-permissive behaviour in place. Where: Most are one line each in your server config, reverse proxy, CDN, or framework headers. (https://owasp.org/www-project-secure-headers/)

For each item, tell me exactly which file to create or which code to change, and where it lives in a project using the stack above. If my framework has a native API for the task (e.g. Next.js metadata exports, `app/sitemap.ts`, Astro layouts, Nuxt `useHead`), prefer that over raw HTML or server config.

Full analysis: https://stackscope.dev/launch/je81feyp/safeexam

I own https://safexam.in. StackScope (https://stackscope.dev) analysed it on 3 June 2026 and scored it 7.6/10 against their production-readiness checklist. Please help me apply the fixes below. This is a checklist, not a blocker list; Score-affecting fixes are the ones that move the number, Optional items are nice-to-haves with zero score impact that I can skip for an MVP.

Tech stack StackScope detected on the site (use this when suggesting where each fix lives in the codebase):
cloudflare, cloudflare-dns, cloudflare-email-routing, google-site-verification, hsts, http3, nextjs, react, tailwindcss, vercel

Score-affecting fixes (these improve the StackScope score; [HIGH] = most impact):
- [HIGH] Add the missing security response headers. Missing 5 of 6 standard browser headers. Start with `Referrer-Policy`, `X-Content-Type-Options`, and `X-Frame-Options` (one line each), then roll out CSP in `Report-Only` mode. Why: Each header limits a class of browser-side attack: clickjacking, XSS, MIME sniffing, plaintext fallback. Missing headers leave default-permissive behaviour in place. Where: Most are one line each in your server config, reverse proxy, CDN, or framework headers. (https://owasp.org/www-project-secure-headers/)

Optional best-practice improvements (no score impact):
- [MEDIUM] Move DMARC from p=none to quarantine or reject. DMARC policy is `p=none` (monitoring only). Why: p=none lets receivers log failures but still deliver spoofed mail. Real protection only kicks in at quarantine or reject. Where: After reviewing a few weeks of `rua=` reports and confirming your real senders pass, change to `p=quarantine` (sends spoofs to spam) and then `p=reject` (blocks them outright). (https://dmarc.org/overview/)
- [MEDIUM] Configure DKIM with your email provider. No DKIM record found at the common selectors we check. Why: DKIM signs your outbound mail with a cryptographic key receivers can verify. Without it, receivers can't tell your real mail from a spoof. Where: Ask your email provider for their DKIM setup. It's usually one TXT record at `{selector}._domainkey.{your-domain}`. (https://www.cloudflare.com/learning/dns/dns-records/dns-dkim-record/)
- [LOW] Trim your <title> tag to under 60 characters. Currently 63 characters: "SafeExam | Online Exam Software for Colleges &amp; Universities" Why: Google truncates titles around 60 characters in search results, so anything past that gets cut off mid-sentence and costs click-through rate. Where: Aim for 50-60 characters that include your main keyword and brand. (https://developers.google.com/search/docs/appearance/title-link)
- [LOW] Trim your meta description to under 160 characters. Currently 185 characters: "Affordable exam integrity for institutions: tab monitoring, violation tracking, question banks, and secure online examinations — more than Google Forms, less than enterprise proctoring." Why: Google truncates around 155-160 characters on desktop SERPs (less on mobile), so anything past that won't appear in the snippet. Where: Edit your `<meta name="description">` tag. Put the most click-worthy phrase first. (https://developers.google.com/search/docs/appearance/snippet)
- [LOW] Add /.well-known/security.txt. Why: Lets security researchers report vulnerabilities responsibly through a published contact channel. Where: Create `/.well-known/security.txt` with a `Contact:` email and `Expires:` date. (https://securitytxt.org/)
- [LOW] Add an llms.txt file. Why: Helps AI models understand your site's content and how to use it. Not yet a standard but gaining adoption. Where: Create `/llms.txt` at the site root with a brief overview and key URLs. (https://llmstxt.org/)
- [LOW] Add Link response headers. Why: Lets agents discover your sitemap, privacy policy, and docs without parsing HTML, which most lightweight agents skip. Where: Set `Link:` response headers in your server config or framework middleware. (https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Link)
- [LOW] Add MTA-STS. No MTA-STS DNS record published. Why: Prevents mail to your domain being downgraded to plaintext mid-flight by a network attacker. Most launches don't have this, so deploying it puts you a tier above generic email-security checks. Where: Publish a TXT record at `_mta-sts.{your-domain}` plus a policy file at `https://mta-sts.{your-domain}/.well-known/mta-sts.txt`. (https://datatracker.ietf.org/doc/html/rfc8461)
- [LOW] Add a TLS-RPT record. No TLS-RPT record at `_smtp._tls.{domain}`. Why: Receivers can tell you when STARTTLS handshakes to your mail server fail. Without it, silent TLS failures are invisible. Where: Publish one TXT record at `_smtp._tls.{your-domain}` like `v=TLSRPTv1; rua=mailto:tls-reports@yourdomain.com`. (https://datatracker.ietf.org/doc/html/rfc8460)

For each item, tell me exactly which file to create or which code to change, and where it lives in a project using the stack above. If my framework has a native API for the task (e.g. Next.js metadata exports, `app/sitemap.ts`, Astro layouts, Nuxt `useHead`), prefer that over raw HTML or server config.

Full analysis: https://stackscope.dev/launch/je81feyp/safeexam

Using an autonomous agent?

Point the agent at this SKILL.md URL and ask it to follow the skill. The framing stops agents defaulting to an open-ended page review.

https://stackscope.dev/launch/je81feyp/skill.md

Share your score

Your score card renders automatically when you share the link.

Or embed a badge

Two badge options. Pick whichever fits your story.

Current score

Shows the latest score and updates within a few minutes of any recrawl. Best for ongoing display: if you fix something and recrawl, the badge reflects the new score automatically.

<a href="https://stackscope.dev/launch/je81feyp/safeexam"><img src="https://stackscope.dev/badge/je81feyp/current.svg" alt="StackScope score for SafeExam" height="24" /></a>

Launch score

Pinned to your launch-day snapshot and never changes. Marked with a small gold corner ribbon. Best for press kits, launch retrospectives, or anywhere you want a permanent record of how you shipped.

<a href="https://stackscope.dev/launch/je81feyp/safeexam"><img src="https://stackscope.dev/badge/je81feyp.svg" alt="StackScope launch score for SafeExam" height="24" /></a>

Using a Content-Security-Policy? Both badges are <img> tags from our domain, so your CSP needs to allow them. Add stackscope.dev to your img-src directive (example: img-src 'self' stackscope.dev;). Without it, browsers silently block the badge and visitors see a broken image.