txtllms — AI search, control, visibility

About

Why txtllms exists

What this tool is for, who built it, and how it works.

1. The problem

People increasingly get answers from ChatGPT, Perplexity, Google's AI Overviews and Claude rather than from a page of blue links. When one of those systems summarises your industry, it decides on its own which pages of your site matter, what they say, and whether you get named at all.

Websites already have a file for telling crawlers what to do: robots.txt. But robots.txt only says yes or no to a path. It has no way to say "here is what this site is, here are the pages that actually explain it, read these first." That gap is what llms.txt, an open specification, was proposed to fill.

2. What this tool does

txtllms.com generates a first draft of an llms.txt file for your website. You give it a domain; it returns a structured, human-readable map of that site - a short description, the sections that matter, and the URLs worth reading - in the format the specification describes. You then review it, edit anything you disagree with, and publish it at yourdomain.com/llms.txt.

The point is to save the boring part. Writing one of these by hand for a site with a few hundred pages means opening the sitemap, deciding what matters, and formatting the result. That is maybe an afternoon. This does the first pass in about a minute so you can spend your time on the judgement calls instead.

This is a starting point, not a finished file. The draft is generated automatically from what your sitemap exposes. Read it before you publish it - you know which of your pages matter and the generator does not.

3. How it works

  1. You submit a domain and an email address. The email is where the finished file is sent - there is no account to create and no password to set.
  2. We read your public sitemap. The tool fetches your robots.txt, follows the sitemap references it finds there (or checks the usual locations if there are none), and collects up to 250 page URLs. It reads nothing behind a login and nothing your own robots.txt excludes.
  3. That URL list is turned into a draft. The list is passed to Google's Gemini API, which groups the pages and writes the file in llms.txt format.
  4. The file arrives in your inbox as a .txt attachment, ready to review and upload to your site's root directory.

Exactly what data is involved at each step, and who processes it, is set out in the Privacy Policy.

4. What it cannot do

It is worth being blunt about the limits, because plenty of tools in this space are not.

What it does do is make your site's structure and purpose legible in one place, in a format that is cheap for a model to read. That is a reasonable bet, not a guarantee.

5. Who is behind it

txtllms.com is built and run by Garvit Malhan, an individual trading as txtllms.com, based in India. It is an independent project.

We are not affiliated with, endorsed by, or acting on behalf of the authors of the llms.txt specification, OpenAI, Google, Anthropic, Perplexity, or any third-party directory of llms.txt files. Where the site links to those, the links are references, not endorsements.

6. What it costs

Nothing. The generator is free to use, there is no paid tier, no trial, and no card is collected anywhere on this site. We ask for an email address because that is how the file is delivered.

Submitting the generator form does not subscribe you to anything. The newsletter is a separate, optional sign-up with an unsubscribe link in every message. We also offer an optional free consultation call; it is general information, not professional advice, and nothing is sold on it.

7. Contact

A real person reads this mailbox. Questions, bug reports, corrections, data requests and takedown requests all go to the same place:

We aim to reply within two business days. Your use of the site is governed by the Terms of Service and the Privacy Policy.