Skip to content

Free tool

Is your website ready for AI agents?

ChatGPT, Claude and Perplexity read websites and cite sources. The Agent Ready Check tests in seconds whether your website is readable for AI systems: llms.txt, robots.txt, AI crawler rules, structured data and Markdown for agents.

What gets checked

15 checks, four areas

Discoverability

  • robots.txt present and readable
  • Sitemap linked or at /sitemap.xml
  • llms.txt in the correct format

Bot access

  • Site reachable without a bot challenge
  • AI crawlers (GPTBot, ClaudeBot, PerplexityBot and others) not locked out
  • DNS health: DNSSEC consistent (broken DNSSEC makes the site unreachable for many resolvers)

Content & structure

  • Structured data (JSON-LD/Schema.org)
  • Title and meta description
  • Exactly one H1 plus lang attribute
  • Markdown for agents (Accept: text/markdown)

Protocols (bonus checks, no points deducted)

  • MCP discovery at /.well-known/mcp.json
  • Content Signals in robots.txt
  • Web Bot Auth (signature directory)
  • DNS-AID (TXT record at _aid)
  • OAuth and API catalog discovery under /.well-known/

Why this matters

A growing share of buying decisions no longer starts at Google but in a chat window. AI systems answer questions directly and link the sources their answer was built from. If you are not readable there, you do not exist - no matter how good your content is.

The good news: the technical base is manageable. Clean crawler rules, an llms.txt, structured data and machine-readable formats take a few hours to set up. That is exactly what this tool checks - and for every issue it shows you how to fix it.

We wrote an in-depth guide on the most-checked item: llms.txt: what it actually does (and doesn't).

FAQ

Frequently asked questions

What does agent-ready mean?

AI systems like ChatGPT, Claude and Perplexity visit websites, read content and cite sources in their answers. Agent-ready means your website is accessible and machine-readable for these systems - crawlers are allowed in, the structure is unambiguous, and machine-friendly formats like llms.txt or Markdown versions of your pages exist.

What is llms.txt?

llms.txt is a Markdown file at the root of your website (like robots.txt, but for AI systems). It contains a title, a short description and curated links to your most important pages. The standard was proposed by Answer.AI in 2024 and is used by a growing number of tools and documentation platforms.

Does llms.txt actually do anything today?

Honest answer: Google has confirmed it does not currently use llms.txt, and OpenAI documents no official support either. But some AI tools and crawlers already read the file, adoption is growing, and it takes about half an hour to set up. Minimal investment, zero risk - which is why we include it in the check.

Does my robots.txt also block AI answers?

If your robots.txt locks out crawlers like GPTBot, ClaudeBot or PerplexityBot, those systems cannot read your content - and therefore cannot cite you as a source. That can be a deliberate choice (content protection), but it costs visibility in AI answers. The check shows you which rule currently applies.

What does the check do technically?

The check fetches your homepage, robots.txt, llms.txt and sitemap live, and additionally tests whether your website serves a Markdown version for the request header Accept: text/markdown. Only the public origin is checked, and no results are stored permanently.

Is being agent-ready enough to show up in ChatGPT?

No. The technical base ensures AI systems can read you - whether they cite you is decided by content, mentions and authority. Agent-ready is the prerequisite, not the guarantee. For everything after that, we are happy to talk about your content and structure.