CrawlRadar - AI SEO Checker & GEO Audit

Score any page 0–100 for AI search: can ChatGPT, Perplexity, Gemini & Google AI Overviews cite it? Prioritized fixes, 100% local.

Open any page and see whether AI answer engines can actually find it, read it, and quote it — with a 0–100 score and a prioritized list of what to fix. Search is shifting from ten blue links to synthesized answers. Most sites are still built for browsers and crawlers, not for the models that now sit between you and your readers. CrawlRadar tells you where a page stands, on the page you're already looking at. HOW IT WORKS Click the icon, press Scan. In about a second you get: • A 0–100 citation-readiness score and a letter grade • Blockers — the issues that make a page uncitable no matter how good the writing is • A ranked fix list, biggest win first, each with the points it recovers and the effort it takes • A full breakdown of all 11 checks, each one explaining what it found and why it matters • A shareable score card WHAT IT CHECKS Retrieved — can an AI reach and parse this page? (30 points) • AI crawler access: whether robots.txt lets GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot and 7 more actually fetch this URL • Content in initial HTML: whether the text exists before JavaScript runs, since most AI crawlers don't execute it • Indexable and accessible: noindex directives, canonical sanity, paywalls and sign-in walls • llms.txt: present and well-formed Selected for synthesis — is it extractable and worth quoting? (70 points) • Answer-first structure: does each section lead with its answer, or bury it behind setup? • Chunk extractability: are paragraphs self-contained enough to quote out of context? • Heading hierarchy: one H1, no skipped levels, descriptive and question-shaped headings • Scannable formatting: lists, tables, paragraph length, emphasised key terms • Citable specifics: density of numbers, dates, prices and named entities — AI quotes specifics, not adjectives • Q&A structure: question headings and FAQPage schema, the most citable shape on the web • Structured data: JSON-LD present, correctly typed, and parsing cleanly Most of the weight sits on extractability rather than technical crawlability. That's deliberate — it's the part that actually decides whether a passage gets quoted, and the part generic SEO tools handle worst. FIX IT WITH YOUR AI CODING AGENT Press "Copy fix prompt" and CrawlRadar writes a complete brief for Claude Code, Cursor, or any coding agent with access to your site. It's not generic advice. The prompt carries the evidence behind every verdict — the exact section headings that bury their answer, the crawlers named in your robots.txt, your measured specifics density — plus copy-pasteable JSON-LD and robots.txt snippets. It orders the work blockers-first, tells the agent not to change your meaning or keyword-stuff, and lists what already passes so nothing gets regressed. Scan, copy, paste, and your page gets fixed. PRIVACY — READ THIS PART CrawlRadar collects nothing. Not anonymised, not aggregated, not "for analytics". • No account, no login, no sign-up wall • No servers. There is no backend to send anything to • No analytics, no telemetry, no tracking, no ads • Nothing is stored, on your machine or anywhere else • Nothing runs on any page until you click Scan The extension requests exactly two permissions: activeTab and scripting. It has no host permissions, which is why Chrome shows no "read your data on all websites" warning when you install it. The scanner is injected only on your click, reads the page, computes the score locally, and is gone. The only network requests are to the site you're already on, for its own /robots.txt, /llms.txt, and HTML — the same files your browser already fetches. WHO IT'S FOR Founders, bloggers, indie hackers, content marketers, technical writers and SEO consultants who publish things and want them cited. Especially useful if you own the site and can act on what you find. HONEST ABOUT WHAT WE DON'T KNOW No tool can promise you citations. The answer engines don't publish their ranking criteria, and anyone claiming otherwise is guessing. CrawlRadar scores what's observable and well-evidenced: whether crawlers can reach you, whether the content survives extraction, and whether it's structured the way answer engines demonstrably prefer to quote. Where a signal is unproven, it says so — the llms.txt check literally tells you it's a quick win with low proven impact rather than pretending it's a ranking factor. Checks that measure rather than infer say so. Checks that guess say that too. Free, with no paid tier and no upsell in this version.