Can AI agents use your site?

Paste a web address. We read it the way an AI assistant does and tell you what works, what is broken, and how to fix it. Free, no account.

Checking https://orbilo.co/webmcp/playground/fixtures/none

Reading it the way an agent does, against both standards. Up to 45 seconds.

What this checks

Three answers, one run. Each gets its own grade, because they answer different questions.

UCP

Can an agent find what you sell and how to buy it?

  • Profile reachable where agents look, without redirects
  • HTTPS, content type, and required caching headers
  • Required fields, versioning, schema ownership
  • Nothing secret accidentally made public

WebMCP

Can an agent act on your site, not just read it?

  • Whether your page registers tools, and via the current entry point
  • Every tool's name, description, and parameter schema
  • Every tool declaration we can find in your page's code
See a live example

Access

Can agents reach your site at all?

  • robots.txt rules for GPTBot, ClaudeBot, PerplexityBot and the rest
  • Whether the catch-all group locks everyone out
  • Crawl-delay settings that slow agents to a crawl
  • Training-only crawlers reported separately, never graded down

New to this? Start with what UCP is, what WebMCP is, or read why agent readiness matters.