llms.txt Generator
Generate a spec-shaped llms.txt from your sitemap — with an honest note about what it's worth.
The URL you enter is fetched by our server to run the checks. Nothing you paste is stored.
About llms.txt Generator
llms.txt is a proposed convention: a Markdown file at the root of your site that gives a language model a curated map of your important pages, with a one-line summary of each. The format is simple — an H1 with your site name, a blockquote summary, then `##` sections of `- [title](url): notes`.
This generator builds one from what you already publish. It reads your robots.txt for `Sitemap:` declarations, falls back to `/sitemap.xml`, fetches up to 50 pages, and takes the title and meta description from each. You get a file shaped to the spec that you can edit down — the curation is the part worth your time, and no generator can do it for you.
Read the honesty panel above the output before you spend an afternoon on this. Google has publicly documented llms.txt as having no effect on its systems. No major vendor has committed to reading one. Measured citation lift in published tests is nil. We build the tool because people ask for it and because the file costs an hour and harms nothing — not because we think it moves the needle.
If you want to spend that hour on something that measurably matters, spend it on server-rendering your content so a crawler that doesn't run JavaScript can read it, or on making sure you haven't accidentally blocked the search crawlers in robots.txt. Both are checked by the audit, and both have evidence behind them that llms.txt does not.
Frequently asked
›Does anyone actually read llms.txt?
No engine has committed to it. Google has said it has no effect on its systems, and there is no published evidence of citation lift from adding one. The convention has real momentum among documentation sites, and some tooling consumes it, but treat it as optional housekeeping rather than an optimisation.
›What is the difference between llms.txt and llms-full.txt?
llms.txt is the index — links and one-line summaries. llms-full.txt is the convention for shipping the full text of those pages in one file, which is mostly useful for documentation sites that want to be pasteable into a context window in one go.
›Is llms.txt a replacement for robots.txt or sitemap.xml?
No. robots.txt controls access and is honoured today. sitemap.xml controls discovery and is honoured today. llms.txt is a curation hint that may be honoured someday. If you only fix one file, it is not this one.
›How many pages should I list?
Fewer than you think. The point is curation — the twenty pages you'd actually want quoted, grouped into sections. A dump of every URL in your sitemap adds nothing a sitemap didn't already say.
›What if my site has no sitemap?
The tool tells you plainly and stops. It won't crawl your site to discover URLs — that would be a much heavier request against someone else's server than a free tool has any business making.