# What Is llms.txt — and Should Your WordPress Site Add One in 2026?

> A straight answer for WordPress owners: what llms.txt is, whether anything actually reads it yet, and when it is worth adding in 2026.

*Source: https://plugins.viglan.com/what-is-llms-txt-wordpress-2026/*

If you’ve seen **llms.txt** mentioned and wondered whether it’s something your WordPress site needs, here’s a straight answer with the 2026 evidence attached — not the hype version.

## The one-line definition

llms.txt is a plain Markdown file at the root of your domain that gives large language models a clean, curated map of your site. It sits at `https://yoursite.com/llms.txt`, the same way robots.txt and your XML sitemap do. The difference is the audience: robots.txt tells crawlers where they may go, a sitemap lists every URL for search engines, and llms.txt hands an AI model a short, human-readable summary of the pages that actually matter.

## What goes in it

The [specification](https://llmstxt.org/) is deliberately simple. A single H1 with your site or project name. An optional blockquote with a one-sentence summary. Then H2 sections — “Docs”, “Products”, “Blog” — each a list of links with a short note explaining what each one is. There’s a special *Optional* section for links a model can skip when it’s short on context. Two extensions round it out: **llms-full.txt**, which inlines all that content into one document, and per-page **.md** copies that strip away theme markup.

## Does anything actually read it? (The honest picture)

This is where you deserve candour. As of mid-2026:

- **Google Search does not use llms.txt.** John Mueller has confirmed it more than once.
- **OpenAI and Anthropic** point site owners to robots.txt for controlling AI crawlers; neither documents llms.txt as a production input.
- Adoption studies put the file on around **10% of surveyed domains**, and one large crawl found the overwhelming majority of llms.txt files received **zero direct AI requests** in a given month.

Where it genuinely earns its place today is **AI developer tools**. Assistants such as Cursor, GitHub Copilot and Claude fetch documentation on demand, and a tidy llms.txt steers them to the right pages while wasting far less of their context window. If you run docs, an API, or a product people build against, that alone can justify the file.

## llms.txt vs robots.txt vs sitemap.xml

They’re complements, not substitutes. **robots.txt** is permission — who may crawl what. **sitemap.xml** is inventory — every URL, for search engines. **llms.txt** is curation — the short list of what matters, written to be read. Adding one doesn’t change the other two.

## So should you add one?

Our take: yes, if it’s automated — and no, if it means hand-editing a Markdown file forever. The cost of serving llms.txt is near zero and the downside is nil, so the insurance argument is reasonable while the convention settles. But a static file you write by hand drifts out of date the moment you publish your next post, and a stale map is worse than none. The value is only there if the file regenerates itself from your real content.

That’s exactly the gap [AI Guide](https://plugins.viglan.com/plugins/ai-guide-by-mhz/) fills for WordPress: it builds a spec-compliant llms.txt, an expanded llms-full.txt, and clean Markdown copies of your pages automatically, and keeps them current as your site changes — all from settings, with everything switched on at maximum coverage by default.