---
title: "The llms.txt format and standard, explained"
url: https://hostmy.blog/docs/rankready/llms-txt-format-explained/
date: 2026-08-25
modified: 2026-08-25
lang: en
author: "Aditya Sharma"
description: "Understand the llms.txt format: its Markdown structure, index versus full file, honest status, and how RankReady builds it from your content."
word_count: 709
---

# The llms.txt format and standard, explained

The llms.txt format is a plain Markdown file published at your site root. It gives AI clients a clean, structured map of your most important pages. An H1 names the site, an optional blockquote summarises it, and H2 sections list Markdown links with short descriptions. It is a proposed convention, not a ratified standard.

## How the file is structured

llms.txt is Markdown, and it reads top to bottom in a fixed order. Once you know the four parts, you can look at anything RankReady generates and judge whether it says what you want.

- **H1: the site name.** A single top-level heading, the name a reader would recognise. There is exactly one per file.
- **An optional blockquote: the summary.** One short line, prefixed with `>`, that states what the site is. This is the sentence an AI client sees first, so it carries weight.
- **H2 sections: your grouped links.** Each section is a heading like Docs, Blog, or About, and under it a bullet list. Every bullet is a Markdown link followed by a one-line description of what that page covers.
- **An optional "Optional" section.** A section literally titled Optional signals lower-priority links that a client may skip when it is working within a tight budget.

A minimal file looks like this:

`# Acme Docs

> Documentation and guides for the Acme platform.

## Docs

- [Getting started](https://acme.example/start.md): install and first run
- [API reference](https://acme.example/api.md): endpoints and parameters

## Optional

- [Changelog](https://acme.example/changelog.md): release history
`

The links point at [Markdown version](https://hostmy.blog/docs/rankready/wordpress-markdown-for-ai/)s of your pages where those exist. That is why RankReady serves a distinct `.md` URL for every post: the index can point a client straight at clean text instead of a full HTML page.

## llms.txt versus llms-full.txt

These are two files with different jobs, and RankReady serves both.

- `/llms.txt` is the **index**. It is short, it lists and describes your pages, and it lets a client decide what to fetch next.
- `/llms-full.txt` is the **full corpus**. It concatenates the actual content of your pages into one file, so a client can read the whole site in a single request without following any links.

The index is the safer default to publish widely. The full file is larger and better suited to clients that want everything at once. In RankReady the full file is off by default and the index is on, which matches how most owners want to start.

## Where the standard actually stands

Be clear-eyed about this. llms.txt is a proposed convention that gained traction across 2024 and 2025. It is not enforced by any browser, search engine, or governing body, and no client is obliged to read it. Some AI tools fetch it, some ignore it, and that support can change without notice.

What publishing the file does is concrete and modest. It puts your chosen pages at a predictable address in a clean form, so any client that does look knows where to find a structured map. It does not compel a client to read your site, quote it, or rank it anywhere. A fetch of the file tells you it was retrieved, nothing more. Treat llms.txt as making your content easy to discover, not as a lever on any outcome.

## How RankReady builds it from your content

RankReady assembles the file from your WordPress content automatically. It fills the H1 and the summary blockquote from your [Brand Identity](https://hostmy.blog/docs/rankready/brand-terms-setup/), then builds the H2 sections from your published pages and posts, writing each link and its one-line description for you. You do not hand-edit a text file.

Power users shape the output from **AI Visibility → Brand Identity**. The site name and the one-line summary you set there are exactly what appear at the top of `/llms.txt` and `/llms-full.txt`, so this is where you control the first thing a client reads. The About field feeds the same surfaces. Adjust those fields, save, and the served file updates.

The endpoint responds to conditional requests as well: `/llms.txt` supports ETag and returns a 304 when a client sends a matching `If-None-Match`, so repeat fetches stay cheap.

## Related reading

- **[llms.txt example](https://hostmy.blog/docs/rankready/llms-txt-examples/)s**: annotated files from real sites, so you can compare structures before shaping your own.
- **How to [add llms.txt](https://hostmy.blog/docs/rankready/how-to-add-llms-txt-wordpress/) to WordPress**: the step-by-step for enabling and serving the file with RankReady.