An AI skill is a structured system prompt or set of custom instructions that gives AI assistants like ChatGPT, Claude, and Gemini specialized capabilities. It includes a clear role definition, specific knowledge areas, step-by-step processes, and quality criteria that help AI perform specific professional tasks.

How does Skill This work?

Simply describe your professional expertise in plain text. Our free AI skill generator analyzes your input and transforms it into a structured system prompt. The skill is then graded on clarity, specificity, and usefulness. You can share your skill via a unique URL or copy it to use as custom instructions with any AI assistant like ChatGPT, Claude, or Gemini.

Is Skill This free to use?

Yes, Skill This is completely free. No account or payment is required. You can create unlimited skills and share them with anyone.

How is my skill graded?

Skills are graded on a scale from A+ to F (0-100 points). The grade reflects how clear, specific, and actionable your skill is. Higher grades indicate skills that AI assistants can use more effectively.

How do I use my skill with an AI assistant?

Copy your generated skill and paste it into the system prompt or custom instructions of any AI assistant. For ChatGPT, use Personalization > Custom Instructions. For Claude, add it to a Project's instructions. For Cursor or other IDEs, save it as a .cursorrules or CLAUDE.md file. The AI will then adopt that skill and perform tasks according to your expertise.

AI Skill Report Card

Building South African Web Experiences

A87·Sep 8, 2026·Source: Web

Quick Start14 / 15

When asked to audit or build a South African-facing site, run through tiers in order — don't skip to Tier 3 features while Tier 1 gaps exist.

Priority check (do this first):
1. Page weight per route < 2 MB? (Lighthouse / WebPageTest, throttled to 3G)
2. LCP < 2.5s, CLS < 0.1, INP < 200ms on mid-tier Android emulation?
3. SSL valid + POPIA privacy notice present + no third-party trackers leaking PII?
4. WCAG 2.1 AA: alt text, 4.5:1 contrast, keyboard nav?
5. "Call Now" / "WhatsApp Us" CTA visible above the fold on mobile?

If any of these fail, fix them before touching personalization, PWAs, or analytics.

Recommendation▾

Add a third example showing a Tier 2/3 scenario (e.g., PWA rollout) to demonstrate range across all tiers, not just Tier 1 audits.

Workflow15 / 15

Progress:

Tier 1 audit: mobile-first, performance, SEO+AEO, security/POPIA, accessibility, CTAs
Tier 1 fixes implemented and re-tested
Tier 2 audit: personalization, local integrations, schema, interactive content, analytics, trust signals
Tier 2 enhancements scoped and prioritized by conversion impact
Tier 3 roadmap: PWA, content automation, AI security monitoring
Final report with before/after metrics

Tier 1 — Core Essentials (non-negotiable)

Mobile-First Optimization: Design for smallest viewport first. Thumb-reachable nav (bottom or sticky), no hover-dependent interactions, tap targets ≥44px. Test on throttled 3G/4G, not just WiFi.
Performance & Load Speed: Target <2 MB total page weight. Compress/lazy-load images (WebP/AVIF), defer non-critical JS, inline critical CSS. Measure Core Web Vitals (LCP, CLS, INP) via Lighthouse/PageSpeed Insights/CrUX.
SEO + AEO: Use local keywords (city/township/service combos), structured headings, and content phrased as direct answers to questions (for AI search engines like Google SGE, Perplexity, ChatGPT browsing). Add FAQ sections answering real customer queries verbatim.
Security & POPIA Compliance: Enforce HTTPS everywhere, secure form endpoints, cookie consent banner, and a plain-language privacy notice stating what data is collected, why, and for how long. Avoid unnecessary third-party scripts that leak PII.
Accessibility (WCAG 2.1 AA): Alt text on all images, semantic HTML, minimum 4.5:1 contrast, visible focus states, screen-reader-tested forms.
Conversion-Optimized CTAs: Persistent "Call Now" and "WhatsApp Us" buttons (using tel: and wa.me links) above the fold and sticky on scroll. A/B test copy, color, and placement.

Tier 2 — Advanced Growth Drivers

Personalization Engine: Adjust hero content/offers based on geolocation (city/province), referral source (UTM/ad source), and device type. Keep logic client-light to avoid hurting performance.
Local Integrations: PayFast/Ozow/SnapScan checkout, WhatsApp Business API for order/support flows. Always provide a fallback (EFT/manual) for integration failures.
Structured Data & Schema: Implement LocalBusiness, FAQPage, and AggregateRating JSON-LD. Validate with Google's Rich Results Test before shipping.
Interactive & Video Content: Compress video (H.264/AV1, <5 MB for short clips), provide calculators/quote tools as lightweight JS widgets, make content easy to share via WhatsApp (short links + preview images).
AI-Driven Analytics & Insights: Track funnel drop-off points, use predictive/behavioral tools (e.g., GA4 + BigQuery, Hotjar) to flag friction. Report actionable insights, not raw dashboards.
Trust & Reputation Signals: Embed real reviews with Review/AggregateRating schema. Never fabricate ratings — validate against actual review platform data.

Tier 3 — Future-Ready Enhancements

Data-Light PWAs: Add service worker for offline caching of key pages, app manifest for "Add to Home Screen," and graceful degradation on poor connections.
Content Freshness & Automation: Use AI-assisted workflows to keep FAQs, pricing, and service pages current; flag stale content (>90 days) for review rather than auto-publishing unchecked.
AI-Enhanced Security Monitoring: Set up anomaly detection for traffic spikes, brute-force login attempts, and phishing-style form submissions; alert rather than silently block, to avoid false positives hurting real users.

Recommendation▾

Include a 'bad output' example contrasting a superficial audit (e.g., just adding a cookie banner) with the correct thorough approach, per the pitfalls listed.

Examples16 / 20

Example 1: Input: "Audit this ISP's landing page for prepaid data customers in Gauteng." Output:

Page weight: 4.8 MB (fail, target <2 MB) — largest offenders: unoptimized hero image (2.1 MB), unminified JS bundle (900 KB)
LCP: 4.2s on throttled 4G (fail, target <2.5s)
No POPIA privacy notice found (fail)
WhatsApp CTA missing on mobile view (fail)
Recommendation: compress hero to WebP (~150 KB), defer JS, add sticky WhatsApp button, add privacy notice footer link — re-test before Tier 2 work.

Example 2: Input: "Add trust signals to a township-based retailer's site." Output: Embed 5 real Google Reviews with Review schema, add AggregateRating JSON-LD, display "Verified by Google" badge linking to live review page, ensure star rating markup validates in Rich Results Test.

Recommendation▾

Description is solid but slightly long — could be tightened while still covering what/when triggers.

Best Practices

Always test on throttled connections (Slow 3G/Fast 3G) and mid-tier Android devices — not just desktop WiFi.
Fix Tier 1 issues completely before investing in Tier 2/3 — a fast, compliant, accessible site converts better than a feature-rich slow one.
Prefer tel: and wa.me links over contact forms for South African users — WhatsApp and calls convert far better than email.
Validate all schema markup before deployment; broken structured data can hurt SEO instead of helping.
Keep personalization and analytics scripts lightweight — don't let Tier 2 features regress Tier 1 performance budgets.

Common Pitfalls

Adding PWA/offline features before basic page weight is under control — this masks the real problem.
Using stock hero videos/images without compression, blowing past the 2 MB budget.
Fabricating or inflating review scores — always link schema to real, verifiable review data.
Treating POPIA compliance as a checkbox (just adding a cookie banner) without an actual plain-language privacy notice.
Over-personalizing based on assumptions (e.g., city-based pricing) without testing that it doesn't confuse or alienate users.
Silently auto-publishing AI-generated content updates without human review — leads to stale or incorrect info going live unchecked.