Is your site ready for AI agents?
The first scanner with 28 checks from the 2026 agent-protocol stack — MCP, A2A, Agent Skills, API Catalog, Web Bot Auth, Content-Signal. See how ChatGPT, Claude, Perplexity and agent runtimes read your site.
AI-readiness checker verifies site preparedness for AI search (ChatGPT, Perplexity, Claude, Gemini): 15+ AI bot accessibility (GPTBot, ClaudeBot, PerplexityBot), llms.txt presence, structured answers, TL;DR blocks, FAQ/HowTo schema, author bylines. Returns a grade and recommendations.
Want a weekly re-check of this?
Drop your email — we will re-run this check every 7 days and alert you if anything degrades (SSL expiry, DNS change, header regression). Free.
One-click unsubscribe in every email. We never share email addresses. By subscribing you agree to our privacy policy.
What is under the hood
How it works
Fetch the site and its service files: llms.txt, robots.txt, meta tags, JSON-LD
Run 20 checks: LLM content, crawler access, agent protocols
Show your AI-Score and what to fix first
Is your site ready for AI search?
ChatGPT, Perplexity, Claude and AI search engines read sites differently from classic crawlers. AI-Check runs 20 checks — from llms.txt and robots rules to agent protocols — and shows what keeps your site out of AI answers.
LLM content (5 checks)
llms.txt and its structure, full-content link, markdown content negotiation, Content-Signals, Speakable markup.
AI crawler access (5 checks)
robots.txt rules for GPTBot, ClaudeBot, PerplexityBot; X-Robots-Tag, AI meta tags, sitemap and Link headers.
Agent protocols (8 checks)
agents.json, ai-plugin, MCP server card, A2A agent card, agent skills, api-catalog, OpenAPI, OAuth discovery — readiness for AI agents.
Citability and AI-Score
Authorship signals and Web Bot Auth, plus a 0–100 score with a prioritized fix-first list.
Who uses this
SEO specialists
visibility in AI Overviews and answer engines
Content teams
citations in ChatGPT and Perplexity
Developers
llms.txt, JSON-LD, MCP and agent protocols
Business owners
customers increasingly arrive from AI answers
Common Mistakes
Best Practices
Frequently Asked Questions
What does AI-Check test?
The tool runs 20 automated checks: LLM content (llms.txt, markdown negotiation, Speakable), AI-crawler access (GPTBot, ClaudeBot, PerplexityBot rules in robots.txt, X-Robots-Tag, meta tags), 8 agent protocols (agents.json, MCP, A2A, OpenAPI) and citability signals. You get an AI-Score plus a prioritized fix list.
What is llms.txt and is it required?
llms.txt is a root-level file that summarizes your site structure and key pages for language models — a robots.txt analogue for LLMs. It is not formally required, but it helps AI systems understand your site quickly and improves the odds of being cited correctly in answers.
Should I allow GPTBot and ClaudeBot in robots.txt?
If you want your site cited in ChatGPT, Claude and Perplexity answers — yes: a blocked crawler cannot read your content, so there is nothing to cite. Blocking makes sense only if you deliberately want to stay out of AI answers and training data.
How do AI crawlers differ from Googlebot?
Many AI crawlers do not execute JavaScript: they read raw HTML, service files and structured markup first. For them llms.txt, JSON-LD and clean HTML structure matter most — not behavioral factors. That is why a site with solid classic SEO can still be invisible to AI systems.
Does AI readiness affect classic SEO?
Not directly: llms.txt and agent protocols are not Google or Yandex ranking factors. Indirectly — yes: structured markup, clean HTML and clear metadata improve classic indexing too. GEO builds on top of solid SEO, not instead of it.
How often should I re-check my site?
Right after any robots.txt, markup or structure change — immediately; otherwise monthly is enough: AI-access standards (llms.txt, Content-Signals, agent protocols) evolve fast, and yesterday's green check can go stale. You can also enable a weekly automatic e-mail re-check.
Related guides
Longer-form reading on this topic from the knowledge base.