Free tool
llms.txt Generator
llms.txt is a markdown file that tells AI systems which pages on your site are worth reading first. Fill in your site details, add your key pages, and get a standards-compliant file you can copy or download — everything runs in your browser, nothing is uploaded.
Generator
Create your llms.txt
Type your site name, base URL, and the pages an AI should read first. The preview updates as you type.
# Your Site
Save as UTF-8 plain markdown at your root path: /llms.txt. A longer companion can live at /llms-full.txt.
Now check whether AI actually cites your site.
Run a free AI visibility check ->Background
What is llms.txt?
A table of contents for machines, not humans.
llms.txt is a community proposal, published at llmstxt.org, for a single markdown file at the root of your website. Where robots.txt tells crawlers what they may not read, llms.txt tells an AI system what it should read: the handful of pages that explain what your product is, how it works, what it costs, and how to get started.
The motivation is context windows. An AI system answering a question about your product cannot ingest your whole site, so the proposal gives it a curated map instead — short enough to load into a prompt, structured enough to parse mechanically.
One honesty note before you invest time here: llms.txt is not an official standard. Major AI platforms have not formally committed to reading it, and publishing one guarantees nothing about crawling, citation, or ranking. It is cheap, harmless hygiene — the measurable work of AI visibility still lives in crawlable pages, quotable content, and third-party mentions.
Spec
llms.txt format rules
Plain markdown, a strict skeleton, no front matter.
The file is UTF-8 markdown served at /llms.txt. The structure is deliberately rigid so any tool can parse it:
1. An H1 with your site or product name. This is the only required element. 2. A blockquote directly after it — one or two sentences summarizing what the site is and who it serves. 3. H2 sections, each holding a bullet list of links in the form - [Label](URL): short description. The description is optional but recommended; keep it under a sentence. 4. A final section named exactly ## Optional for secondary links an AI may skip when context is tight.
A complete, minimal example:
# Acme Analytics > Acme Analytics is a privacy-first web analytics tool for indie SaaS teams. ## Docs - [Quickstart](https://acmeanalytics.com/docs/quickstart): Install the tracker in five minutes - [API Reference](https://acmeanalytics.com/api): Endpoints for events, funnels, and exports ## Product - [Pricing](https://acmeanalytics.com/pricing): Plans, limits, and the free tier - [Changelog](https://acmeanalytics.com/changelog): Recent releases and fixes ## Optional - [Blog](https://acmeanalytics.com/blog): Essays on analytics and privacy
Deploy it as plain text at your root path — for example https://you.com/llms.txt — and keep it under a few dozen links. If you maintain full documentation in the same spirit, the proposal reserves /llms-full.txt for an unabridged companion. Regenerate the file whenever your key pages move; a stale map is worse than none.
Context
Why llms.txt matters for AI search
Small file, honest expectations.
AI search engines — ChatGPT Search, Perplexity, Google AI Overviews — assemble answers from retrieved pages under tight context budgets. The pages that get cited are the ones that state plainly what a product does and for whom. An llms.txt file is your chance to hand those systems exactly that set of pages, in exactly that wording, instead of leaving the selection to chance.
Adoption is real but early. A growing number of documentation sites and dev-tool products publish llms.txt files, and several AI tooling vendors read them when they exist. What you should not expect: a citation bump the week you ship one. There is no public evidence that llms.txt alone changes whether ChatGPT or Perplexity recommends you, and anyone promising that is selling something.
The reliable loop is: publish the file, make sure the pages it points to are crawlable and quotable, then measure whether AI answers actually mention you. That last step is what Answerlume is for — run buyer-style prompts on ChatGPT Search and Perplexity and read the raw answers with their citations. The GEO checklist covers the other 19 things worth doing alongside this one.
FAQ
What is llms.txt?
llms.txt is a community proposal (llmstxt.org) for a markdown file at your site root that lists the pages an AI system should read first: product pages, docs, pricing, comparisons. It is written in plain markdown with an H1 site name, a blockquote summary, and link lists grouped under H2 sections.
Is llms.txt an official standard? Will it make AI cite my site?
No. llms.txt is an experimental community proposal. Major AI platforms have not formally adopted it as a standard, and publishing one does not guarantee crawling, citation, or ranking. Treat it as low-cost hygiene alongside the work that demonstrably moves AI visibility: crawlable pages, quotable content, and third-party mentions.
Where do I put the llms.txt file?
Serve it as UTF-8 plain markdown at your site root, e.g. https://you.com/llms.txt. If you maintain a longer, unabridged version, the proposal reserves /llms-full.txt for it — the short file stays a concise map, the full file can carry complete documentation.
What should I include in my llms.txt?
Keep it short and current. Start with your site name as the H1 and one sentence describing what the product does. Then link the pages that explain you best: product overview, docs quickstart, pricing, comparisons, changelog, and support. Put secondary material under an ## Optional section. A stale or bloated llms.txt is worse than none — regenerate it when your pages change.
File generated? Next question.
Is AI actually citing your site yet?
Run a free AI visibility check ->Three free checks per website: buyer-style prompts on ChatGPT Search and Perplexity, with raw answers and citation URLs. Then work through the GEO checklist.