llms.txt for your website: what we found on 38 Indian sites, and what to put in yours

There is a reel going around that says "add llms.txt or ChatGPT will ignore you". You have probably seen some version of it. So this week I went and looked at what Indian businesses actually have at that address.

The answer surprised me. Most of the files were there. Very few of them said anything useful.

Full disclosure, we have one too. We added an llms.txt file to our own site on 11 October. So I am not here to call it useless. I am here to tell you what it does, and what it does not.

What is llms.txt?

It is a plain text file at yoursite.com/llms.txt, written in Markdown. The idea is to give AI tools a short, clean summary of your site: who you are, what you sell, and links to the pages that matter.

It is a proposal. Nobody has to follow it. The llms.txt site describes it as "a proposal to standardise on using an /llms.txt file to provide information to help agents use a website". It was published in September 2024 and updated in August 2026.

What did we find on 38 Indian websites?

On 28 October 2026 we opened the websites of the 40 businesses in our daily teardowns of 22 to 28 October 2026. 38 loaded: 16 coaches or educators and 22 D2C brands.

So 24 of 38 sites "have llms.txt". That number tells you very little.

The brand files

All 22 brands run on Shopify. 21 of them serve the same file, headed "Agent Instructions", which tells an AI agent how to browse the store and shop on the buyer's behalf. It is the same text with the brand name swapped in. Nobody at these brands wrote it, and it says nothing about what makes them different.

One luggage brand had a real file: a paragraph on what they make, where they are based, when they started and their price band. That is the version I would want.

The coach files

Two coaches had llms.txt, both made by a WordPress plugin that lists every page on the site. One file, for a coach whose site title calls him a business coach, opens with a post called "Hello world!", the sample post WordPress creates on day one, and a page called "Event Settings Test". If an AI reads that file, the opening thing it learns is a test page.

The other 14 coaches had nothing at all.

Does llms.txt help you show up in ChatGPT or Google?

Here is the honest answer.

Google's own guide on AI features is blunt. It says you don't need to create new machine readable files, AI text files, or markup to appear in AI Overviews or AI Mode. A page needs to be indexed and eligible to show with a snippet. That is all.

For ChatGPT, OpenAI's crawler page talks about robots.txt settings and does not mention llms.txt. I could not find a public statement from either company that they rank or cite sites because of this file.

So why bother? Because it is cheap, it cannot hurt, and AI agents that browse sites, like coding tools and research assistants, do read it when it is there. Think of it as a clean front desk note. It will not win you clients on its own.

What matters more than llms.txt?

Three things, in this order.

Your robots.txt

This decides if AI crawlers can read your site at all. 3 of the 38 sites blocked a list of AI crawlers, including GPTBot and Google Extended. All 3 used the exact same list, which looks like a template or hosting setting the owners may not know about.

OpenAI says GPTBot is for training and OAI-SearchBot decides if your pages can appear in ChatGPT search. Those 3 sites did not block OAI-SearchBot, so they can still show up in ChatGPT search. Check yours. Open yoursite.com/robots.txt and read every "Disallow: /" line.

Your About page and schema

A clear About page with your name, role, city and offer, plus Person or Organization schema. Our Person schema guide has code you can copy.

Pages that answer buyer questions

When we asked ChatGPT 24 coaching questions with search on, it cited coaches' own websites again and again, the ones with clear pages about their programs. Read how coaches get recommended by ChatGPT for that data.

What should your llms.txt say?

Write it by hand. It takes 20 minutes. Here is a template for a coach. Change every line.

# Priya Mehta, online fitness coach

> Priya Mehta is an online strength coach for working women over 35, based in Pune, India. Coaching since 2019. Programs from ₹6,000 to ₹24,000 for 12 weeks. Sessions in English and Hindi.

## Programs
- [12 week Strong at 40 program](https://www.priyamehta.in/strong-at-40/): live group coaching, 3 sessions a week, ₹24,000
- [Home workout plan](https://www.priyamehta.in/home-plan/): self paced, ₹6,000

## About and proof
- [About Priya](https://www.priyamehta.in/about/): certifications, story, photos
- [Client results](https://www.priyamehta.in/results/): with dates and permission

## Contact
- [Book a free 15 minute call](https://www.priyamehta.in/book/)
- WhatsApp: +91 98XXX XXXXX

The rules I follow:

• Facts a stranger can check. Name, city, year, prices, languages.
• Links to real pages, never to test pages, tags or sample posts.
• Keep it short. If it runs past one screen, cut.
• Update it when your prices change.

We keep an llms.txt on our own site, written by hand, and a longer llms-full.txt with the text of our guides. You can open ours to see a real one.

Should you use a plugin to make it?

You can, but read the result before you leave it live. The two coach files in our check show what happens when nobody does. If your plugin lists every page, remove the junk and keep the 5 to 10 pages that sell. A plugin can write the file. It cannot decide what matters in your business. That part is yours.

If you want us to look at your robots.txt, llms.txt, schema and About page in one pass, start with the free Site Scan. The deeper Agent Ready Audit, with a blind test on 3 AI tools, is ₹9,999 and is described on our funnel audit page.

Sources

• The Human Algorithm, llms.txt and robots.txt check, 28 October 2026. Websites of the 40 businesses in our daily teardowns of 22 to 28 October 2026, 38 loaded (16 coaches or educators, 22 D2C brands). Files read from /llms.txt and /robots.txt. Names kept private.
• llms.txt, The /llms.txt file, v2, read 28 October 2026 (published 3 September 2024, modified 10 August 2026).
• Google Search Central, AI features and your website, read 28 October 2026.
• OpenAI, Overview of OpenAI crawlers, read 28 October 2026.