# Sonicar AI Score: Is your site ready for search, answer engines and AI agents? Sonicar AI Score is a free audit by Sonicar Tech. Paste a website or docs URL. In about 20 seconds you get three scores — SEO, AEO and Agent-Ready Docs — a prioritized fix brief, and a report you can share with your team. ## What does the audit check? ### SEO: Can search engines crawl, index and rank it? Technical and on-page SEO: indexability, metadata, headings, structured data and delivery speed. - **Crawlability & Indexing** (8 checks): Search engines can reach, crawl and index the page. - **Titles & Metadata** (6 checks): Title, description, language and mobile metadata are present and well-sized. - **Content & Headings** (6 checks): Clear heading structure, enough depth, accessible images and healthy internal links. - **Rich Results & Sharing** (3 checks): Structured data and social cards that power rich results and link previews. - **Speed & Delivery** (5 checks): Fast server response, lean HTML, compression and secure assets. ### AEO: Will ChatGPT, Perplexity, Claude and AI Overviews cite it? Answer Engine Optimization: AI crawler access, entity markup, extractable answer-ready content and trust signals. - **AI Crawler Access** (3 checks): Answer-engine crawlers are allowed by robots.txt and not blocked by your firewall or CDN. - **Structured Data for AI** (5 checks): Entity and page-type markup that tells models who you are and what the page is. - **Content Extractability** (4 checks): The content is in the HTML, in semantic containers, with deep-linkable sections. - **Answer-Ready Content** (5 checks): Question-led headings, a direct lead answer, scannable structure and quotable facts. - **Authority & Freshness** (3 checks): Authorship, update dates and trust pages that make content safe to cite. - **AI Discovery Files** (4 checks): llms.txt, markdown alternates and sitemaps that help AI systems find content. ### Agent-Ready Docs: Can AI agents consume your docs cleanly? The Agent-Friendly Docs checks: llms.txt, markdown delivery, context-window size, structure, stability and access. - **Content Discoverability** (6 checks): Agents can find your docs index (llms.txt), follow its links, and every page points back to it. - **Markdown Availability** (2 checks): Clean markdown through .md URLs or content negotiation instead of bloated HTML. - **Page Size and Truncation Risk** (4 checks): Pages fit agent context windows and the useful content starts early. - **Content Structure** (3 checks): Tabs, headings and code fences stay parseable after an agent serializes the page. - **URL Stability and Redirects** (2 checks): URLs resolve cleanly with correct status codes and predictable redirects. - **Observability and Content Health** (3 checks): llms.txt coverage, markdown/HTML parity and cache headers keep agent-facing content accurate. - **Authentication and Access** (2 checks): Agents can read the docs without hitting login walls, or have a public alternative. ## What do you get? Every failing check comes with the evidence we saw and the exact fix. Hand it to a writer, a developer or straight to your coding agent. - **What to fix first:** Issues ranked by the points they cost you, across all three pillars. - **AI agent prompt:** One click copies a prompt for Claude Code, Cursor or Copilot to fix it in your repo. - **PDF and markdown:** Export a clean PDF for stakeholders or the full report as markdown. - **Shareable link:** Every report has its own URL with a branded preview card for Slack and LinkedIn. ## See how other sites score The latest listed result for each site. Open any report to see every check, or re-run it to measure progress release over release. https://score.sonicar.tech/#audits ## How every audit is scored Each run fetches your site directly — the page, robots.txt, sitemaps, llms.txt and a deterministic sample of pages as HTML, markdown and as AI crawlers — then applies the same weighted rules every time, so you can compare releases and competitors fairly. Scores are weighted: high-impact checks count 3×, medium 2×, low 1×. A warning earns half credit and checks that don't apply are skipped. The overall score is the average of the three pillars. Grades: A+ 97+, A 90+, B 80+, C 70+, D 60+, F below 60. ## Questions teams ask us ### What is AEO (answer engine optimization)? AEO is the practice of making your content easy for AI answer engines — ChatGPT search, Perplexity, Claude, Microsoft Copilot and Google AI Overviews — to find, understand, trust and cite. It builds on SEO but adds AI crawler access, entity markup, answer-first writing and machine-readable files like llms.txt. ### How is AEO different from SEO? SEO optimizes for ranking in a list of links. AEO optimizes for being quoted inside a generated answer. The same technical foundation matters, but AEO also depends on whether AI crawlers are allowed through your robots.txt and firewall, whether content is visible without JavaScript, and whether pages lead with concise, quotable answers. ### What does agent-ready documentation mean? AI coding agents such as Claude Code, Cursor and Copilot read documentation directly. Agent-ready docs publish an llms.txt index, serve clean markdown via .md URLs or content negotiation, keep pages within context-window limits, avoid login walls and return correct status codes. Our Agent-Ready Docs pillar follows the Agent-Friendly Documentation checks used across the industry. ### What is llms.txt? llms.txt is a markdown file at the root of your site that gives AI models a curated index of your most important pages: an H1 title, a one-line summary in a blockquote, and H2 sections listing links with short descriptions. Linking each entry to a markdown version of the page makes it even easier for agents to use. ### Does it work for Arabic and bilingual websites? Yes. The audit recognizes Arabic question headings, Arabic-Indic digits and Arabic "About" and "Contact" pages, checks that Arabic pages declare dir="rtl", and validates hreflang on bilingual sites. Every report is available in English and Arabic. ### Can I audit a site that blocks bots? If a firewall returns 403 to our auditor, we tell you instead of producing a misleading score. That result is itself a signal: the same firewall usually blocks AI crawlers. Allowlist verified AI crawlers or run the audit on a public page. ### Is the audit free, and what do you store? Yes, it's free with no signup. We store the report so you can share its link, and you choose whether it appears in the public recent-audits list. We only fetch public pages. Sonicar Tech can turn the findings into a fix plan if you want help. --- Run an audit at https://score.sonicar.tech/ · Docs index for AI agents: https://score.sonicar.tech/llms.txt · Built by Sonicar Tech (https://sonicar.tech)