agentleFS
viknesh20-20viknesh20-20/claude-code-tool-kit/.claude/skills/web-launch-check/SKILL.md

web-launch-check

Skill7 starsChanged 7 months agoMITView on GitHub

Pre-launch audit for any website. Verifies privacy policy, ToS, cookie consent, sessions, sitemap, robots.txt, SEO meta tags, structured data, accessibility (WCAG AA), Core Web Vitals, and compliance frameworks (GDPR/CCPA). Returns severity-graded findings with a GO / NO-GO verdict.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 10 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run this before any production website launch to catch missing legal pages, broken SEO, accessibility violations, or compliance gaps that …

    line 8
  2. Distinctive:

    Before pushing a marketing site, landing page, or web app to production.

    line 12
  3. Distinctive:

    After a major redesign that changed page structure, routes, or third-party integrations.

    line 13
  4. Distinctive:

    Before adding analytics or tracking scripts.

    line 14
  5. Distinctive:

    When taking over an existing site to assess current state.

    line 15
  6. Very specific:

    Every page has unique <title> (≤ 60 chars).

    line 47
  7. Very specific:

    Every page has <link rel="canonical">.

    line 49
  8. Very specific:

    Every page has Open Graph tags (og:title, og:description, og:image, og:url, og:type).

    line 50
  9. Distinctive:

    Every page has Twitter Card tags.

    line 51
  10. Very specific:

    No PII echoed in URLs (?email=...).

    line 95
  11. Distinctive:

    Read the build output if available.

    line 108
  12. Very specific:

    Check each layer above; mark each finding Critical | High | Medium | Low | Pass.

    line 109
  13. Distinctive:

    Always check every layer. Don't skip layers because the project "is just a landing page."

    line 144
  14. Distinctive:

    Don't let style/preference issues block the verdict — only Critical and High block.

    line 149
  15. Distinctive:

    Don't recommend frameworks.

    line 150
  16. Very specific:

    Don't paste full legal text — point to templates (see /legal-pages skill).

    line 151

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/web-launch-check

Run this before any production website launch to catch missing legal pages, broken SEO, accessibility violations, or compliance gaps that would expose the user to legal risk or hurt search ranking.

When to use

  • Before pushing a marketing site, landing page, or web app to production.
  • After a major redesign that changed page structure, routes, or third-party integrations.
  • Before adding analytics or tracking scripts.
  • When taking over an existing site to assess current state.

What it checks

Layer 1 — Legal pages (CRITICAL — block launch)

PagePath checkedRequired when
Privacy Policy/privacy, /privacy-policy, footer linkAlways (any data collection, including cookies/IP)
Terms of Service/terms, /tos, /terms-of-serviceAny signup, transaction, or user-generated content
Cookie Policy/cookies, section in /privacyEU/UK visitors, or any non-essential cookies
Contact / Address/contact, footerEU jurisdictions (Impressum, EU consumer law); always recommended
Refund / Cancellation/refunds, /cancellationAny e-commerce or paid product

Layer 2 — Consent & tracking (HIGH)

  • Cookie consent banner present before any non-essential cookie is set.
  • Reject button as prominent as Accept.
  • Tracking scripts (Google Analytics, Meta Pixel, etc.) blocked until consent.
  • "Do Not Sell or Share" link if California users (CCPA/CPRA).
  • Global Privacy Control signal honored.

Layer 3 — Sessions & auth (HIGH)

  • Cookies have Secure, HttpOnly, SameSite=Lax (or Strict).
  • Auth tokens not in localStorage if long-lived.
  • Logout invalidates session server-side.
  • CSRF protection on state-changing requests.
  • Session expiry explicit; idle timeout for sensitive flows.

Layer 4 — SEO essentials (HIGH)

  • Every page has unique <title> (≤ 60 chars).
  • Every page has unique <meta name="description"> (~150 chars).
  • Every page has <link rel="canonical">.
  • Every page has Open Graph tags (og:title, og:description, og:image, og:url, og:type).
  • Every page has Twitter Card tags.
  • Structured data (JSON-LD) where relevant (Organization, Article, Product, BreadcrumbList, FAQ).
  • <html lang> set.
  • Exactly one <h1> per page; heading hierarchy correct.
  • Alt text on every meaningful image.

Layer 5 — Site-wide SEO files (HIGH)

  • robots.txt at root with explicit rules and sitemap reference.
  • sitemap.xml at root, all canonical URLs, lastmod dates.
  • /.well-known/security.txt (recommended).
  • 404 page returns HTTP 404.
  • 500 page exists for server errors.

Layer 6 — Performance / Core Web Vitals (HIGH)

  • LCP ≤ 2.5s on 75th percentile mobile.
  • INP ≤ 200ms.
  • CLS ≤ 0.1.
  • Total blocking JS ≤ 250 KB gzipped (or justified).

Layer 7 — Accessibility (WCAG 2.1 AA) (HIGH)

  • Keyboard reachable everywhere.
  • Focus visible.
  • Color contrast ≥ 4.5:1 (text), ≥ 3:1 (large text and UI).
  • Form labels associated.
  • Error messages with aria-describedby.
  • prefers-reduced-motion respected.
  • Skip-to-content link present.

Layer 8 — Compliance frameworks (situational)

  • GDPR (EU users): lawful basis documented, DSAR endpoint, 72-hour breach notification plan.
  • CCPA/CPRA (California): "Do Not Sell or Share" link.
  • HIPAA (health data): BAAs with vendors, encryption verified.
  • PCI-DSS (cards): no raw card numbers handled; tokenization via Stripe/Adyen.

Layer 9 — Forms & PII (HIGH)

  • HTTPS-only.
  • Server-side validation on every form.
  • Rate limiting on signup, login, contact, password reset.
  • CAPTCHA on public forms.
  • No PII echoed in URLs (?email=...).
  • Email verification before privileged actions.

Layer 10 — Email deliverability (MEDIUM, if applicable)

  • DMARC / DKIM / SPF set up.
  • Unsubscribe link in every marketing email.
  • Physical address in marketing email footer (CAN-SPAM).

How the check runs

  1. Detect the framework (Next.js, Astro, SvelteKit, Nuxt, plain HTML, etc.) by reading the project structure.
  2. Walk the routes / pages — collect what exists and what's referenced.
  3. Read the build output if available; otherwise inspect source.
  4. Check each layer above; mark each finding Critical | High | Medium | Low | Pass.
  5. Suggest a fix snippet for each failure.

Output format

## Verdict: GO | CONDITIONAL (fix Critical + High) | NO-GO

## Executive summary
2–3 sentences. Most important risk first.

## Findings by layer

### Layer 1 — Legal pages
- [Critical] /privacy missing — add a route and a stub policy.
- [High]    /terms missing — needed because there's a signup form at /signup.

### Layer 4 — SEO essentials
- [High]    No <meta name="description"> on /pricing, /features, /blog/*.
- [Medium]  Open Graph tags missing site-wide.

### Layer 7 — Accessibility
- [High]    Color contrast 3.2:1 on primary CTA — needs ≥ 4.5:1.
…

## Quick wins
3–5 items the user can ship in one PR.

## Defer-with-issue
Items not blocking launch but tracked.

Do / don't

Do:

  • Always check every layer. Don't skip layers because the project "is just a landing page."
  • Cite the exact file and line for each finding.
  • Provide a fix snippet for every High and Critical finding.

Don't:

  • Don't let style/preference issues block the verdict — only Critical and High block.
  • Don't recommend frameworks; work with what's there.
  • Don't paste full legal text — point to templates (see /legal-pages skill).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.