Free Tool
Is your site ready for AI agents?
Audit any site for the agent web. We check that AI agents can reach it, read it without running JavaScript, and use its machine signals like llms.txt and structured data, then grade the result. No signup needed.
Reads the page and its agent files. Takes a few seconds.
How it works
Three steps, no signup.
No crawlers, no cached data. Every check is a live request to the site you name, read the way an agent would read it.
Enter any URL
Type a website address and run the check. We fetch the live page and its agent-facing files, with no crawlers and no cached data in the way.
We audit it the way an agent would
We read the raw HTML without running JavaScript, the way most agents do, parse robots.txt for AI-crawler rules, and look for llms.txt, structured data, a sitemap, and an MCP endpoint.
Read the grade
You get an A-to-F grade across three areas, access, readability, and agent signals, with a pass, warn, or fail on every check so you know exactly what to improve.
What does "ready for AI agents" actually mean?
It comes down to three things. Access: can an agent reach your page (HTTPS, no CAPTCHA or bot wall) and are the major AI crawlers allowed in your robots.txt. Readability: can an agent that does not run JavaScript read your content, plus headings, metadata, and structured data. Agent signals: do you publish the machine-readable files agents increasingly look for, like llms.txt, a sitemap, feeds, and an MCP endpoint. We grade all three and roll them into a single score.
Why does content without JavaScript matter so much?
Most AI crawlers and agents fetch your raw HTML and do not execute JavaScript. If your site renders its content client-side, a single-page app that ships an empty shell, those agents see a blank page. We measure the real text in your raw HTML, so a low score here is the clearest signal that agents cannot read you.
Should I allow or block AI crawlers?
That is your call, and this tool reports it factually rather than judging it. If you want your content to appear in AI answers and be usable by agents, allow the major AI crawlers in robots.txt. If you would rather not be trained on or summarized, blocking them is legitimate, just know it means "not ready" by this tool's definition.
What are llms.txt and MCP?
llms.txt is an emerging plain-Markdown file at your site root that gives agents a curated map of your most useful pages. MCP (Model Context Protocol) is a standard for exposing tools and data that agents can call directly, so they can act on your site rather than just read it. Both are early, so we treat them as bonuses that lift your score rather than penalties. Sentinel publishes an llms.txt and ships its own MCP server.
Built for the agent web.
Sentinel monitors your site around the clock and exposes that monitoring to agents through its own MCP server, so both you and your tools know the moment something breaks.