Single Site Gen

Project spec

Fill in what you have. Empty fields get smart defaults. The generator reuses every best-practice rule our analyzers audit for — so the prompt you paste into Claude produces code that would already pass Mega Analyzer, Site Analyzer, E-E-A-T Audit, AI Citation Readiness, Mobile Parity, and Hosting Health out of the gate.

Best-practice modules to bake in

Every module is ON by default. Toggle off any you don't need and the prompt shrinks accordingly.

What this prompt tells the LLM to do

This generator is tuned against the same rulebook our audit tools score against. The output prompt encodes every check so the first draft would already score high across every bucket:

• SEO foundations — title ≤ 70, meta 150–160, single H1, canonical, HTTPS, lang, viewport (zoom allowed — WCAG 1.4.4), charset, no meta noindex, no X-Robots-Tag: noindex.
• Crawl validation — robots.txt served, no Disallow conflict for User-agent: *, Sitemap: declaration, sitemap.xml with valid <urlset>, every page in the sitemap, internal out-links from every page (no dead ends).
• Schema — JSON-LD Organization (or subtype), Person, BreadcrumbList, FAQPage, Speakable, Article rules (text|image|video, datePublished timezone, author.url), WebSite with SearchAction, Service for each offering.
• E-E-A-T — Person schema with jobTitle/knowsAbout/alumniOf/worksFor, Organization, ≥3 sameAs including Wikidata/ORCID where applicable, rel="me" identity links.
• AI search readiness — llms.txt, llms-full.txt, .well-known/agent-card.json, .well-known/ai-plugin.json, AI crawler rules (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot, Applebot-Extended).
• Accessibility (WCAG 2.2 AA) — skip link, semantic landmarks, alt text, ARIA labels, focus-visible styles, 4.5:1 contrast, keyboard operable, reduced-motion respected.
• Performance / Core Web Vitals — HTML ≤ 250 KB, DOM ≤ 1,500 nodes, image width+height, WebP/AVIF via <picture>, lazy-loading, preconnect, font-display: swap, critical CSS inline, LCP < 2.5s, CLS < 0.1.
• Mobile parity — same content in desktop & mobile UAs, H2 overlap ≥ 90%, schema parity, no collapsed-only content.
• Security headers — HSTS, CSP (script-src covers all external scripts), Permissions-Policy, X-Frame-Options, X-Content-Type-Options, Referrer-Policy, SRI on external scripts, rel="noopener".
• Voice / readability — Flesch ≥ 60, grade 8–12, passive voice ≤ 10%, sentence-length stddev 6–12, hedges ≤ 3 per 1K words.
• Answer engine optimization — direct-answer paragraphs after each H2, <details>/<summary> FAQs, definition-style formatting, tables for comparisons.
• Hosting config — Netlify/Vercel/Cloudflare/Apache-specific headers, redirects, and caching for the final deployed stack.