Free tool
llms.txt generator
I built this to write an llms.txt file the way the llmstxt.org proposal lays it out: a name, a short summary, then sections of links with a note on each. Paste your sitemap to start from your real pages, then cut the list down to the ones that matter.
Want to know what the file is before you write one? Read my llms.txt glossary entry. To see whether your site already serves one, run the AI crawler check.
llms.txt is a proposal from llmstxt.org, not an agreed standard. No major AI search engine has confirmed that it reads the file to choose what to cite, so treat it as tidy housekeeping and not a ranking lever. Everything here runs in your browser; nothing you type or paste is sent anywhere.
About the site
The H1. The only part the proposal requires.
One or two sentences an agent needs before it reads anything else. Written as a blockquote.
Optional. Paragraphs and lists that explain the site or how to read the links. Headings are escaped, since an H2 would start a new section.
Optional. Lets you type paths such as /docs/start in the links below.
Import from a sitemap or a list
Paste your sitemap.xml or one URL per line, or choose a file. It is read in your browser and nothing is fetched. Up to 2,000 links per import.
Sections of links
Line format: [Title](address): notesThe section Key pages has no valid links yet, so it is left out.
Your llms.txt
Add a site or project name and the file appears here as you type.
Check before you publish
- Error: Add a site or project name. The H1 name is the only part the proposal requires.
- Warning: Add a short summary. It is the first thing an agent reads after the name.
- Warning: The section Key pages has no valid links yet, so it is left out.
- Warning: Add at least one section of links. The file is valid without one, but the links are what an agent follows.
Where it goes
Upload the file as /llms.txt at the root of your site, served as plain text. A file at a path such as /docs/llms.txt covers only the pages under that path.
About llms-full.txt
llms-full.txt is not part of the llmstxt.org proposal. It is a convention some documentation platforms follow: one file with the full text of every page, in Markdown. This tool does not build it, because that needs the content of each page, which only your site has.
Coverage
What it builds and checks
The required name
The H1 with your site or project name. It is the only part the proposal requires, so the file is not ready until you add it.
A summary blockquote
The short summary an agent reads first, written as a Markdown blockquote line by line.
Details without headings
Paragraphs and lists are kept as you write them. A heading line is escaped, because an H2 would start a new section by mistake.
Sections of links
Each H2 section holds a list in the form [Title](address): notes. Square brackets in a title become parentheses and parentheses in an address are encoded, so neither can break the link.
Addresses that work
Only full http and https addresses go in. A link with a password, a local host name or another scheme is flagged and left out, and a repeat of a link already listed is removed.
Sitemap import
Reads a urlset sitemap or a plain list of URLs in your browser. A sitemap index lists its child sitemaps for you to open and paste, since the page fetches nothing.
Method
How it works
- 1
Name and describe the site
Add the name, a one or two sentence summary and, if it helps, a few lines on how to read the links.
- 2
Add links or import them
Type links by hand, or paste your sitemap.xml and let the generator group the pages by folder. Titles start from each address.
- 3
Edit the list down
Rename the titles, add a note to each link and remove pages an agent does not need. A short, chosen list reads better than a full sitemap.
- 4
Copy or download
Fix anything the checks flag, then copy the text or download llms.txt and upload it to the root of your site.
Questions
Frequently asked questions
What is llms.txt?
A Markdown file at the root of a site, or under a path such as /docs, that gives a language model a short summary of the site and a chosen list of links to read. It was proposed at llmstxt.org in September 2024 and revised as version 2 in August 2026. It is a proposal, not an agreed standard.
Do ChatGPT, Perplexity or Google use llms.txt?
Not that any of them has confirmed. Google’s guidance for its AI features says you do not need special AI text files to appear in AI Overviews or AI Mode, and I know of no assistant provider that has said it uses llms.txt to choose sources. It is cheap to add, so add one if you like, but never treat it as a ranking lever. Check your server logs to see whether anything fetches it.
What format does the file follow?
The order the proposal sets: an H1 with the site or project name (the only required part), a blockquote summary, optional details in paragraphs or lists with no headings, then H2 sections that each hold a list of links, each written as [Title](address) with optional notes after a colon. This generator writes exactly that and nothing else.
Where do I put the file?
Upload it as /llms.txt at the root of your site, served as plain text. Version 2 of the proposal also allows a file at a path, such as /docs/llms.txt, which covers only the pages under that path; where more than one applies, the most specific one wins.
What is the Optional section for?
It holds secondary links an agent can skip. In version 1 of the proposal it told context tools what to leave out; version 2 dropped that special meaning but still allows the section as a convention. The generator always writes it last.
What about llms-full.txt?
It is not part of the llmstxt.org proposal. Some documentation platforms publish one file with the full Markdown text of every page and call it llms-full.txt. This tool does not build it, because that needs the content of each page, which only your site has.
Should the links point to Markdown versions of my pages?
If you have them, yes. The proposal suggests that pages an agent might need offer a clean Markdown copy at the same address with .md added (page.html.md) or with the extension replaced (page.md). If your site has no Markdown copies, link to the normal pages.
Does anything I type leave my browser?
No. The whole tool runs on this page. The sitemap import reads the XML you paste or choose with your browser’s own parser and makes no request with it. A draft is kept in this browser’s local storage so a reload does not lose your work, and Start over clears it.
More free tools
Related tools
- Schema Markup GeneratorFill a short form and get valid schema markup for an organization, article, FAQ, product or local business.
- robots.txt GeneratorPick which AI crawlers may read your site, one toggle per bot, and get the robots.txt to match.
- AEO AuditCrawl access, indexability, canonicals, schema, llms.txt and speed for your site, each finding with its fix.
Rather hand it over?
I can find the fixes and ship them for you
I run SEO and answer engine optimization as one managed program: the pages, the technical fixes, the markup and the mentions that AI assistants read. You get a plain monthly report on what changed.
See how I run AEO