Free tool

Is your product ready for AI agents?

Enter your site URL. We check the public surfaces that answer engines and AI agents look for: llms.txt, AI-crawler rules, an MCP server card, an OpenAPI description, Organization data and a dated sitemap. The full result shows on this page, no email.

No card, no email. We don't store the files we read.
–
Enter a URL and run the check to see which agent-readiness signals your site has.
How it works

Enter a URL and the tool fetches seven public surfaces on that domain, each one a signal that an AI agent or answer engine uses to find, understand and act on your product. It reports which are present, scores the set out of 100, and tells you what to add for the ones that are missing.

It is a read from the outside, so it sees what a crawler sees, not what a logged-in app exposes to a connected agent. That is the honest limit of any public check.

Methodology

Each signal is fetched directly, with a 10-second timeout and at most three redirects. A missing or blocked file counts as not present. These are the seven signals, and why each one matters to an agent.

llms.txt
A plain-text map at /llms.txt that points language models to your most important pages.
llms-full.txt
The fuller version at /llms-full.txt, with more context for models that want it.
AI crawler access
robots.txt rules that let GPTBot, OAI-SearchBot, ClaudeBot, Claude-SearchBot and PerplexityBot read the site.
MCP server card
A descriptor at /.well-known/mcp/server-card.json so agents can discover and call your product over MCP.
OpenAPI description
A machine-readable /openapi.json so agents can call your API with typed, reliable requests.
Organization structured data
Organization JSON-LD on the home page so engines can resolve your company as an entity.
Dated sitemap
A /sitemap.xml with real per-URL last-modified dates, a true freshness signal for crawlers.
What to do with your result

A low score is a to-do list, not a verdict. Add a llms.txt file, open your robots.txt to AI crawlers, and publish a dated sitemap first, since those are quick and lift how often answer engines can read and cite you.

The harder signals, an MCP server card and an OpenAPI description, let agents actually call your product rather than just read about it. Peeve adds both out of the box: it gives your users a scoped MCP endpoint and serves these surfaces for you. See the other docs audit and support cost tools too.

Common questions

What does the agent-readiness check look at?

The agent-readiness check fetches seven public signals on your domain: /llms.txt, /llms-full.txt, AI-crawler rules in robots.txt (GPTBot, OAI-SearchBot, ClaudeBot, Claude-SearchBot, PerplexityBot), an MCP server card at /.well-known/mcp/server-card.json, an /openapi.json description, Organization JSON-LD on the home page, and dated entries in /sitemap.xml. It scores each as present or missing and returns a weighted score out of 100.

Is the agent-readiness check free?

Yes. The agent-readiness check is completely free and needs no email or account. Enter a URL and the full result, every signal and what to do about it, appears on the page. It is rate-limited to keep it fair, so wait a minute if you run several checks in a row.

Do you store the data from my site?

No. The agent-readiness check reads a handful of public files on your domain to score them in the moment and does not store the files it reads. It only records the domain checked and the score for product analytics, never the page contents.

Give your product an agent surface, not just a page. 14 days free. No card. No call.