LLMs.txt Generator

Want an llms.txt file without learning the format first?

The llms.txt idea is simple: a Markdown file at the root of a domain that gives language models a short summary of the site and a list of pages worth reading. Writing one by hand is not hard, but getting the heading and blockquote structure right means a trip to the spec. This generator fills that structure in for you. You answer five prompts, and it returns the text ready to save, with a button that downloads it under the correct filename. No data leaves your browser.

Step by step

1. Enter the name of the site or brand. The tool places it on the first line after a single # sign.
2. Describe the site in the larger box. This line is prefixed with > so it renders as a quote, which is how the proposal marks the summary.
3. In the topics box, list subject areas with commas between them; each turns into a list item under a Topics heading.
4. Paste the addresses of your key pages, one to a line, into the pages box, and optionally an email address for the Contact section.
5. Press "Generate llms.txt", then either copy the text or press Download and place the saved file in your web root.

What to watch for

Every field is optional as far as the code is concerned, so confirm the first two lines are not blank before saving. Page entries come out as raw addresses, whereas the published proposal expects a Markdown link per page, ideally followed by a colon and a few words of context. Converting them takes a minute. Also note that the Topics and Contact headings are conveniences added here rather than something the format asks for.

Questions people ask

Do Google, ChatGPT or Perplexity read llms.txt?

None of the major answer engines has confirmed that it reads the file, and there is no public evidence that having one changes what they show. Treat it as an optional extra, not a visibility tactic.

Where should the file live?

At the top level of the host, so the address is your domain followed by /llms.txt. A copy tucked into a subfolder will not be found by anything looking for the convention.

Should I list every page on the site?

No. The point is a short, curated guide. Pick the pages that explain what you do, your main documentation or services, and anything a model would need to describe you accurately. Your XML sitemap already covers the full inventory.

Further reading: AGENTS.md vs llms.txt vs llms-full.txt: Which Agent File Does What and The Machine-Readable Web: What robots.txt, llms.txt, Schema, WebMCP and Entity Signals Actually Do in 2026.

The same tool is also published on the static mirror at seoprocheck.github.io, and the rest of the set is on the free tools page.