What is CiteLens?
CiteLens is a free tool that measures how readable and citable your website is to AI systems like ChatGPT, Claude, and Perplexity. You paste a URL; it fetches the page, runs structural checks, and uses Claude to grade your content — returning a score out of 100 across six categories, each with specific fixes you can ship.
It exists for one reason: search behaviour is shifting from search engines to AI assistants, and most websites are structurally invisible to the crawlers that feed them. CiteLens shows you where you stand and what to change.
The six things we check
- Content clarity — is your writing direct and quotable, or buried in vague marketing language?
- Structured data — do you use schema.org, JSON-LD, and OpenGraph so AI can extract entities?
- Semantic structure — a clear heading hierarchy and lists that AI can navigate and quote.
- Entity definition — does your site plainly state what it is, who it serves, and what it does?
- llms.txt presence — the emerging standard: a markdown map at
/llms.txtthat guides AI to your best content. - Factual density — concrete stats, dates, names, and sources that AI systems prefer to cite.
How it works
- You paste any public URL — no login required.
- We fetch the page exactly as an AI crawler sees it — without running JavaScript, because crawlers like GPTBot and ClaudeBot don't run it either.
- Three categories are scored deterministically in code; three are judged by Claude for stable, trustworthy results.
- You get an overall score and two to three specific recommendations per category.
Why AI readability matters
AI assistants increasingly answer questions directly and cite a handful of sources. If your pages are client-rendered, thin on structure, or vague about what you do, AI systems skip you — and your competitors get cited instead. Three shifts make this urgent:
- ChatGPT, Claude, and Perplexity now cite their sources inline in answers.
- AI crawlers generally do not execute JavaScript, so client-rendered sites are often invisible to them.
- The llms.txt standard, proposed in 2024, lets a site hand AI a curated markdown map of its most important content.
The standards behind AI readability
AI readability builds on concrete, datable standards and named crawlers — the specifics CiteLens checks for:
- llms.txtwas proposed by Jeremy Howard, co-founder of Answer.AI, in September 2024. It is a markdown file at a domain's root that lists a site's key pages for AI systems.
- schema.org launched in 2011 as a joint project of Google, Microsoft Bing, Yahoo, and Yandex. Its JSON-LD vocabulary lets crawlers extract entities like Organization, Product, Article, and FAQPage.
- The major AI crawlers identify themselves by user agent: GPTBot (OpenAI), ClaudeBot (Anthropic), and PerplexityBot (Perplexity). None of them execute JavaScript by default.
- CiteLens scores each of its six categories from 0 to 100 and makes a single Claude API call per audit, so a full report typically returns in 10 to 30 seconds.
Who it's for
CiteLens is built for two audiences: marketers and SEO managers at B2B SaaS companies who are watching organic traffic decline and want to understand their AI visibility, and developers who want to integrate AI-readability checks into their own tooling via the POST /api/audit endpoint.
Frequently asked questions
What is Generative Engine Optimization (GEO)?
Generative Engine Optimization is the practice of structuring a website so AI systems like ChatGPT, Claude, and Perplexity can read, understand, and cite it — the AI-era analogue of SEO.
What does CiteLens check?
Six categories: content clarity, structured data, semantic structure, entity definition, llms.txt presence, and factual density. Each is scored 0–100 with specific recommendations.
Is CiteLens free?
Yes. Paste any public URL for a free instant audit. No login is required, and the same audit is available to developers via the API.