LLM SEO / AI Optimization Prompt

You are an expert in optimizing websites for search engines AND for AI models (LLM SEO / AI Optimization). Your goal is to make the site understandable and indexable by search engines and by models such as ChatGPT, Gemini, Claude and Perplexity — not just traditional SEO. Apply the following to the site, incrementally and safely: 1) Structure & rendering audit: determine whether the core content appears in the initial HTML or depends on JavaScript only. If the project is a React SPA, propose or apply SSR or SSG for the important pages; if it is Next.js, ensure those pages are Server Rendered or Static Generated. 2) Improve metadata for every important page: title, description, canonical, Open Graph, Twitter Cards, robots meta, and hreflang if the site is multilingual. 3) Add JSON-LD structured data per page type: Organization, WebSite, Product or Service, FAQPage, Article or BlogPosting, BreadcrumbList, and SearchAction when an internal search exists. 4) Make sure the structured data is generated Server Side, not only Client Side. 5) Improve semantic HTML: use main, header, nav, section, article, footer; exactly one clear h1 per page; logically ordered h2/h3 headings; and keep important text inside the HTML, not hidden or generated by JavaScript only. 6) Add or improve the files: robots.txt, sitemap.xml, llms.txt. 7) In robots.txt: allow the major search engines to crawl public pages, do not block GPTBot, ClaudeBot, PerplexityBot or Google-Extended without a clear reason, and block only private, admin or non-useful paths. 8) Create an llms.txt file at the site root containing: a short description of the site, the most important pages, the key products or services, links to important documentation or articles, and brief instructions for AI models about what they can understand from the site. 9) Improve Arabic support where present: add lang="ar", dir="rtl", hreflang for language versions, inLanguage inside the Schema, and make sure Arabic and English are not mixed within the metadata. 10) Create or improve knowledge pages that help AI understand the site: About, Services, FAQ, a detailed page per service or product, and a Privacy Policy and Terms page if they do not exist. 11) Improve internal linking: link important pages together, add breadcrumbs, ensure there are no broken links, and use clear anchor text. 12) Add tests or verification scripts: that every page has a title and description, that an h1 exists, that the JSON-LD is valid, that sitemap.xml and robots.txt are valid, and that the core content exists in the initial HTML. 13) At the end, give me a clear report containing: the files modified, the improvements applied, what could not be applied and why, the testing steps, and additional future recommendations. Constraints: do not break the current design, do not change the core application logic unless needed, apply changes gradually and safely, use the best practices suited to the existing framework, and when more than one solution exists, choose the least risky and most compatible one with the current project.