What Is llms.txt, and Does Your Website Need One?

llms.txt is a plain Markdown file placed at the root of a website (yoursite.com/llms.txt) that gives AI assistants and agents a short, curated guide to the site: who you are, what you offer and which pages hold the details. It is an open proposal, not an official web standard, and Google says you do not need it to appear in its AI features.

So should a small business bother? For most sites it is a low-cost, low-risk addition that takes under an hour, as long as you treat it as housekeeping rather than a growth tactic. This guide explains what the file is, who reads it, how to write one and when to skip it. Sources were checked on 12 November 2026.

What is llms.txt?

The idea was published by Jeremy Howard in September 2024 on llmstxt.org, and the proposal was revised as "v2" in August 2026. The reasoning is simple: web pages are built for people, with menus, scripts and layout that make clean text hard to extract, and AI models work within limited context. A small, well-organized file gives an agent a direct path to the information it needs.

The proposal also suggests offering clean Markdown copies of important pages (for example, page.md next to page.html) and linking to them from the llms.txt file. That part is optional and mostly used by software documentation sites.

How is llms.txt different from robots.txt and sitemap.xml?

The three files sit side by side at the site root but do different jobs:

• robots.txt tells crawlers what they are allowed to access. It is about permission.
• sitemap.xml lists every page you want search engines to find. It is about completeness.
• llms.txt is a short, hand-picked summary with the most useful links. It is about context, read on demand when an assistant is helping someone.

llms.txt does not block or allow anything. If you want to control whether AI crawlers can use your content, that still happens in robots.txt and in each AI company's published crawler settings.

What goes in an llms.txt file?

The spec on llmstxt.org defines a simple order, which matters because some tools parse it with code:

1. An H1 heading with the name of the business or site. This is the only required part.
2. A blockquote with a one- or two-sentence summary of what you do.
3. Optional paragraphs or lists with key facts, such as service area, how to get in touch and what you do not offer.
4. H2 sections with link lists, each item written as a Markdown link followed by a colon and a short note, for example "Services", "Pricing", "Policies" or "Blog".
5. An "Optional" section for secondary links an assistant can skip when it is short on space.

For a small business, a good file usually fits on one screen: your name, a plain description, your main service pages, your contact page, any return or service policies, and a handful of your best guides.

Who actually reads llms.txt today?

This is where honesty matters, because a lot of marketing around the file oversells it.

• Google Search does not need it. Google's guidance on AI features and your website states that you don't need to create new machine-readable files, AI text files or markup to appear in AI Overviews or AI Mode, and that normal SEO best practices still apply.
• AI companies use it for their own docs. OpenAI, Anthropic and Google's Gemini API documentation each publish an llms.txt file, which we confirmed loads on 12 November 2026. That shows the format is taken seriously for developer documentation.
• Coding agents and assistants with browsing can fetch the file when a user asks about a site. The llmstxt.org proposal notes that the heaviest use so far is software documentation.
• Chrome's Lighthouse now includes an llms.txt check in an "Agentic Browsing" category, according to the llmstxt.org page and public issue reports in the Lighthouse repository on GitHub. A tool checking for the file is not the same as a search engine rewarding it.

What is not public: whether any major AI assistant uses llms.txt when deciding which businesses to recommend or cite. No vendor we checked has said so. Anyone promising that the file will get you into ChatGPT answers is guessing.

Does your website need an llms.txt file?

Think of it in three groups:

• Add one now if you publish software documentation, an API, a product with setup guides, or any large knowledge base. This is the use the format was built for, and agents already look for it.
• Add one when convenient if you are a service business, store or agency with a clear set of key pages. It costs little, keeps your own summary of the business in one place and does no harm.
• Skip it for now if your site has basic problems: pages that are not indexed, thin service pages, missing contact details or a slow mobile experience. Fixing those does far more for both search and AI answers.

If you run WordPress, plugins such as Yoast SEO and AIOSEO can generate the file, and some hosted platforms create one automatically. Check what they produce; an auto-generated list of every page is closer to a sitemap than to the curated guide the proposal describes.

How to create an llms.txt file in five steps

1. List the pages that answer the questions customers ask most: what you do, where you work, how pricing or quoting works, how to contact you, and your policies.
2. Write a two-sentence summary of the business in plain words. Avoid slogans; an assistant needs facts.
3. Arrange the links under a few H2 headings, each with a short note saying what the page covers.
4. Save it as llms.txt in the site root so it loads at yoursite.com/llms.txt as plain text.
5. Test it the way llmstxt.org suggests: give an AI assistant only the file and ask it questions about your business. If the answers are wrong or vague, rewrite the summary and notes.

Keep it current. An llms.txt that lists retired services or old policies is worse than none, because it hands assistants outdated facts in a format that looks authoritative.

What matters more than llms.txt for AI visibility?

The file is one small piece of answer engine optimization. The larger pieces are the same ones that help in search: pages that answer real questions directly, accurate structured data that matches visible text, consistent business details across the web, and a site that crawlers can reach and load quickly. If you want help with the technical side, our SEO services cover crawl access, structured data and site files such as robots.txt, sitemaps and llms.txt.