Paste your site URL and get a ready-to-edit llms.txt draft: site name, description, and your pages grouped into sections with titles and summaries pulled from the pages themselves. Edit it here, then copy or download. Free, no signup.
3 free drafts a day as a guest · unlimited with a free account ·
When this box has URLs, the sitemap is skipped and only these pages are fetched. Paths like /pricing work too.
The homepage is best. We read its title and description for the top of the file.
Up to 40 pages are fetched, shallow pages first, with their titles and meta descriptions. No sitemap? Paste the URLs instead.
Tidy the draft in the editor, then copy it or download llms.txt and upload it to the root of your site.
Three steps, no build tooling required.
llms.txt (plain UTF-8, Markdown inside).https://yourdomain.com/llms.txt: the public/ folder on static hosts and most frameworks, or a route that returns the text with Content-Type: text/plain or text/markdown.# Acme > Acme makes invoicing software for freelancers and small studios. ## Docs - [Getting started](https://acme.com/docs/getting-started): Install Acme and send your first invoice. - [Payments](https://acme.com/docs/payments): Connect Stripe or PayPal and set payment terms. ## Pages - [Pricing](https://acme.com/pricing): Plans from free to agency. - [Changelog](https://acme.com/changelog): What changed, month by month.
llms.txt is a proposed convention (llmstxt.org, September 2024) for a Markdown file served at /llms.txt. It starts with an H1 naming the site, a one-line blockquote describing it, and then sections of links with short descriptions, so a language model or AI agent with a limited context window can find the pages that matter instead of crawling everything. It is a proposal, not a standard: no major vendor has committed to reading it, but it is cheap to publish and increasingly common.
A minimal file is: a line "# Acme" , a line "> Acme makes invoicing software for freelancers.", then "## Docs" followed by lines like "- [Getting started](https://acme.com/docs/start): Install and send your first invoice." Sections can be anything you like; "## Optional" is the one named in the proposal for pages a model can skip. The generator produces exactly this shape, with sections named after the first path segment of your URLs.
It reads your homepage for the site name (og:site_name or the title) and the meta description, finds your sitemap through robots.txt Sitemap lines or the usual /sitemap.xml, /sitemap_index.xml and /sitemap.txt locations, then fetches up to 40 pages (shallow ones first) with a few requests in parallel and takes each page title and meta description. The whole run stops after about 20 seconds; pages that did not finish are listed so you can add them by hand.
Yes. When no sitemap is found the tool opens a box where you paste the page URLs you want listed, one per line. Only URLs on the same host are accepted, up to 40. Looking for a sitemap does not count against your daily limit, so the retry is free.
At the root of your site, so it is reachable at https://yourdomain.com/llms.txt, served as text/plain or text/markdown. Most static hosts take a file in the public folder; frameworks like Next.js, Astro and Rails need a route or a public file. Reference it nowhere in particular: tools that look for it request the fixed path. Check it with our AI crawler checker, which reports whether the file exists and what its first heading is.
The proposal describes an optional llms-full.txt that contains the complete text of your documentation in one file. It is useful when your content is short enough to fit in a model context and your pages are hard to fetch individually. This tool produces the index file only; a help center platform that renders Markdown can usually generate the full file automatically.
No. robots.txt tells crawlers what they may fetch, sitemap.xml lists every URL for search engines, and llms.txt is a curated guide for language models. AI search crawlers still need robots.txt access and readable server-side HTML to use your pages at all; llms.txt only helps them choose.
Yes. Guests can generate 3 drafts a day without an account and a free Helpdesky account removes the limit. Any public http or https site works; private networks and unusual ports are refused. Nothing is stored.
No signup and no credit card. Each one runs in seconds.
Enter your help center or docs URL and get a scorecard for discoverability, content, structure, freshness and technical health.
Open toolSee which AI crawlers (GPTBot, ChatGPT-User, PerplexityBot, ClaudeBot and more) your robots.txt allows or blocks, plus llms.txt and no-JavaScript readability.
Open toolPaste any website URL and get its color palette as hex codes, plus the logo and favicon.
Open toolAudit any URL for on-page SEO issues: meta tags, headings, structured data, links and performance.
Open toolEvery free tool in one place, with more on the way. All of them run without an account.
Browse the hubHelpdesky help centers are server-rendered with sitemaps and llms.txt built in and kept up to date as you publish articles, so AI search engines always see your latest answers. Free plan, no credit card.
Create a free help center