Generative engine optimization
Content that generated answers can cite
- Access
Stage 1: Access
llms.txt generator and validator
Build an llms.txt from a sitemap or a list of URLs, or validate one you already publish. Labelled an experiment.
Free to use. No login, nothing stored.
What it checks
Access- Site name, one-line summary and grouped links with descriptions from each page
- Format of an existing file against the llms.txt proposal
- Status of every linked URL and the file size
Method question for this stage: Can crawlers reach and render every key page?
Run the check
Results
No check run yet
- Info
Generate: the file appears here with Copy and Download
One row per step shows where the site name, summary and each description came from.
- Info
Validate: one row per format rule, then the status of each linked URL
Each row names the rule from the llms.txt proposal it tests.
How we check
The rules this tool tests, and where they come from
A result is a check against a published rule, not a prediction. Every row shows its working.
Rules and sources
Links go to the operator's or standards body's own documentation. We do not link to third-party summaries.
Read more
Next step
Fix what it finds
This check belongs to the Access stage of the Crawl-to-Citation Method. The service below covers the fix; the Diagnostic ranks it against everything else.
Most engagements start with the Search Visibility Diagnostic ($2,500).
It covers all six method stages, so what this tool finds is ranked against everything else before you commit to any work.
Book the diagnosticllms.txt questions
What the file is, what the tool reads, and what no one has shown it does.
Still have questions? Book a callDoes publishing llms.txt get my site into AI answers?
There is no published evidence that it does. Google has stated that its Search systems, including AI features, do not use the file. Other AI vendors have not documented support for it either. Treat the file as a cheap experiment: publish it, keep it accurate, and measure whether anything fetches it from your server logs. We log what we see on our own site in the Lab.
What does the generated file contain?
An H1 with the site name (from og:site_name or the homepage title), a one-line summary from the homepage meta description, then one section per top-level path with a link line per page: the page title, its URL and its meta description or first paragraph. Pages that failed to load, were not HTTP 200, or are legal boilerplate go under Optional for you to review. Edit the result before publishing it; the tool reads what the pages say, it does not write for you.
Which rules does the validator apply?
The ones in the llms.txt proposal at llmstxt.org: a single H1 as the first line, an optional blockquote summary, H2 section headings, list items of the form "- [text](url): description", absolute http or https URLs, and no duplicates. The size row is informational, since the proposal sets no limit. When you give a site address the tool also sends a HEAD request to the first 20 linked URLs and reports each status.
Last reviewed