What is llms.txt and does my website need one?

The file is a plain-text summary of your website, published at yourdomain.com/llms.txt, written for machines rather than for people. It is the same idea as robots.txt, except that robots.txt controls access and llms.txt supplies context.

What goes in it

The convention is Markdown, kept short. A heading with your business name, a blockquote summarising what you do, then sections describing your key facts and linking to your most important pages with a line of explanation each.

The point is not to repeat your website. It is to hand a machine the map, so it spends its budget reading the pages that matter rather than working out which ones those are.

A workable structure:

• Title and summary. Who you are, where, what you sell, in two or three sentences.
• Key facts. Location, markets served, price ranges, credentials. Specific and checkable.
• Pages. Your most important URLs, each with one line saying what it covers.
• Related properties. Other domains and profiles you control, so the entity picture is complete.

Does anything actually read it?

Here is the honest position, and you should be sceptical of anyone who states it more confidently than this. Adoption is partial. Several crawlers look for the file. No major provider has committed to it as a guaranteed input, and none of them publish exactly what they do with it.

So the case for publishing one is not that it is proven to work. The case is:

• It takes about an hour once.
• It has no downside and no maintenance cost beyond updating it when you add pages.
• The exercise of writing it forces you to state your business facts plainly in one place, which surfaces contradictions you did not know you had.
• If adoption broadens, you are already there.

That is a reasonable trade. It is not a growth strategy, and anyone selling it as one is overstating it considerably.

The part that is genuinely useful today

Writing an llms.txt is the cheapest entity-consistency audit available. You have to decide, in one document, what your business is called, where it is, what it sells and at what price. Most companies discover during that exercise that their website, their Google Business Profile and their listings do not agree.

Those contradictions are a real and measurable problem regardless of whether any crawler reads the file, because they lower a model's confidence in you. Fixing them is covered in does schema markup help with AI search.

llms.txt is not robots.txt

If you only do one of the two, audit your robots.txt. Blocking an AI crawler by accident is a real cost happening today. Missing an llms.txt is a missed option. That decision is worked through in should you block AI crawlers.

Where this site's own file lives

You can read the one for this site at aiseomalaysia.com.my/llms.txt. It is deliberately plain. Copy the structure if it is useful.