# Add llms.txt to Your Site

> Add an llms.txt file at your site root that lists your most important pages as a clean, linkable index agents can read before they crawl your site.

## What llms.txt is

llms.txt is a plain-text Markdown file served at your site root (`/llms.txt`) that gives AI agents a curated, linkable index of your most important pages. Proposed by the llmstxt.org project as a discovery-and-declaration file, it is an ordinary Markdown document: an H1 site name, an optional blockquote summary, then sections of bullet links to your key URLs, each with a short description.

## Why it matters for agents

An agent has a limited context budget and pays a Cost of Retrieval for every page it fetches. llms.txt collapses discovery into a single small file: the agent reads one document and knows which pages answer which questions, without parsing layout HTML. It is the Layer-1 declaration of the discovery dimension. A companion `/llms-full.txt` can inline the full text for agents that want everything in one fetch.

## How to implement it

1. Create a Markdown file with an H1 (your site name), an optional `>` blockquote summary, then `##` sections of `- [Title](https://example.com/page): one-line description` links.
2. Serve it as `text/plain` or `text/markdown` at the exact path `https://yoursite/llms.txt` — root only, no redirect chain.
3. Verify it resolves: `curl -sI https://yoursite/llms.txt` returns `200` and the body is valid Markdown with reachable links.

This satisfies the audit check `discoverability.llms_txt` (pass: `/llms.txt` resolves 200 as Markdown with at least one valid link). Verify format details against the primary spec at llmstxt.org at build, then verify the result with the Agent-Readiness Audit.

Related: [the llms.txt spec entry](/protocols/discovery/llms-txt) · [the discoverability dimension](/agent-readiness/discoverability) · [the Agentic Web Lexicon](/glossary) · [audit your site](/services)