Loading…
Loading…
Build and validate an llms.txt file for AI agents
## Optional
Secondary links an agent can skip when it needs a shorter context.
Enter a site name to start. The H1 title is the only required part.
Upload the file to your site root so it loads at https://yoursite.com/llms.txt. Everything runs in your browser.
Need SEO and content templates?
Blog post frameworks, keyword research sheets, and on-page checklists built for real traffic growth. Starting at $4.
Browse SEO TemplatesAn llms.txt file tells an AI agent what your site is and which pages to read first, in a markdown layout proposed at llmstxt.org in September 2024. Writing one by hand is quick, but small mistakes are easy to make: a missing H1, an H3 where only H2 lists belong, a link written as plain text. This generator builds the file from a form and checks the result against the format line by line.
The form maps one field to each part of the format. The site name becomes the H1. The summary is collapsed to one line and written as a "> " blockquote. Details are copied as plain markdown. Each section becomes an H2 followed by a list of "- [title](url): note" items, and the Optional links go last under "## Optional".
While you type, the generator cleans up input that would break markdown links. It escapes square brackets in titles and encodes spaces and parentheses in URLs as %20, %28 and %29. Blank links and sections without links are dropped. A section you name "Optional" is merged into the real Optional section, and two sections with the same name are merged, so the output never repeats a heading.
The validator runs on every change and on any file you paste in. It strips a byte-order mark, accepts Windows line endings, skips fenced code blocks and reports each problem with its line number. Errors break the layout an agent expects. Warnings are allowed by the format but worth fixing: relative URLs, for example, stop resolving once the file is read away from your site.
A developer-docs team publishing /llms.txt that points agents at markdown copies of the quick start, API reference and webhook guide, with the changelog under Optional.
An SEO lead auditing the llms.txt a CMS plugin generated, to catch H3 headings and bare-text list items the plugin wrote.
A SaaS founder writing a first llms.txt from the links already in the docs sidebar, then downloading it and uploading it to the site root.
An open-source maintainer adding an llms.txt to a project site so coding assistants find the install guide and examples before the blog archive.
Scope note: llms.txt is a proposal, not a ratified standard, and each AI product decides for itself whether to fetch it. This tool checks the file layout only. It does not fetch your URLs to confirm they return 200, and it cannot tell whether any crawler or assistant reads your file.
Enter your site or project name. It becomes the H1, the one section the format requires
Write a one- or two-sentence summary for the blockquote and, if needed, a few lines of plain markdown details
Add H2 sections such as Docs or Guides, each with links: title, URL and an optional short note
Put secondary links in the Optional section so agents can skip them when context is tight
Fix anything the validator flags, then copy or download the file and upload it to your site root as /llms.txt
About the llms.txt Generator
llms.txt is a markdown file served at /llms.txt that gives AI agents a short, curated map of a website: what it is and which pages matter. It was proposed on llmstxt.org in September 2024 as a way to help agents use a site at inference time. The llmstxt.org page contrasts it with sitemaps: a sitemap lists every page for search engines, while llms.txt offers a curated overview for LLMs.
Markdown, in a fixed order: an H1 with the site or project name (the only required section), a blockquote with a short summary, zero or more paragraphs or lists with more detail (no headings), then zero or more H2 sections that each hold a list of links written as "- [name](url)", optionally followed by a colon and notes.
By convention, links under an H2 named Optional are secondary. An agent that needs a shorter context can skip them. Put changelogs, long reference pages and background reading there, and keep the pages an agent needs first in your other sections.
At the root of your site, so it loads at https://yoursite.com/llms.txt. The llmstxt.org proposal also allows a file at a sub-path, covering the pages under that path; when several apply, agents should use the most specific one.
The proposal recommends serving a markdown copy of each page at the same URL with .md appended (page.html.md), or with the extension replaced by .md, and linking to those. Plain HTML URLs still work as links; markdown copies are simply cheaper for a model to read.
It parses the file line by line. Errors: no H1 at the top, a second H1, H3-H6 headings, list items that are not "[name](url)" links, invalid URLs and unclosed code fences. Warnings: a missing summary blockquote, relative or non-http URLs, empty or repeated sections, prose inside a link section and duplicate URLs.
No. robots.txt says what crawlers may access and sitemap.xml lists your pages for search engines. llms.txt is a short, hand-picked reading list for language models; the llmstxt.org page says it can complement robots.txt by giving context for allowed content. Keep all three.
No. The file is built and validated in your browser with JavaScript. Nothing you type or paste is uploaded, and no account is needed.
Generate JSON-LD structured data for Google rich results
Open →SEO & ContentPreview how your site looks on Google, Twitter & more
Open →DeveloperWrite Markdown with instant live preview and HTML export
Open →SEO & ContentFancy text in 24 Unicode styles — copy & paste anywhere
Open →