Scan to tool
Each row is a finding as the scan report words it, and the tool that resolves it.
Read a file or a site and see what is really there.
Previews are illustrations.
Find out whether the file on your site was written for your brand or shipped by your platform.
See where you and another site differ, check by check.
Build the fix from your own choices or your own site.
Previews are illustrations.
Pick the crawlers you allow and copy the file.
JSON-LD prefilled from your own site, ready to paste.
The scan finds the gaps. The tools close them, one at a time.
Previews are illustrations.
Read a site the way an AI crawler would and get the fix for every finding.
The scanner, the fix generators and the playbooks also run as tools inside Claude, Cursor and other MCP clients.
Questions
Yes. Every tool on this page is free and works without an account.
No. There are no accounts and no email collection. Open a tool and use it.
Nothing about what you enter. The robots.txt generator runs in your browser and sends nothing. The llms.txt checker, the schema generator and the comparison send the domain you type to our server, which reads public files from that domain and returns the result without keeping it. To enforce a rate limit, the server keeps a one-way hash of your IP address with a request counter, which expires within a day. The privacy policy has the detail.
The scan reads a site and scores it, so it finds what is missing. The tools act on one finding at a time: check a file, generate the fix, compare two sites. Scan first, then open the tool that fits the finding.
No. They measure readiness: whether crawlers can reach a site and whether its files and markup give a model something to read. None of them can show whether an assistant names you in an answer.