Glossary

What is llms.txt?

A proposed markdown file at /llms.txt that curates a site's key content for AI systems. Widely adopted, rarely fetched — know both facts.

Definition

llms.txt is a proposed standard (Jeremy Howard, 2024): a markdown file at your domain root listing your site's purpose and most important pages, so AI systems can find the good stuff without crawling everything. A companion, llms-full.txt, inlines full content.

Adoption is real — 8.7% of the top 1,000 sites publish one (Rankability, June 2026); ~10% across 300k domains (SE Ranking) — but consumption isn't: Ahrefs found 97% of llms.txt files received zero bot requests in May 2026, and no major LLM provider has committed to using it.

Why it matters

The honest case: llms.txt won't currently boost AI search citations — GPTBot, ClaudeBot, and PerplexityBot overwhelmingly skip it. But it costs minutes, carries no risk, and has one proven consumer: coding agents and IDE tools (Cursor, Claude Code, Copilot) fetch it on documentation sites. Treat it as cheap insurance plus agent-readable docs — and never let it substitute for fixing crawlability, which bots actually use.

Frequently asked

Should I create an llms.txt file?

Yes, if it takes ten minutes — Seeqly's free generator does exactly that. Just hold honest expectations: it's forward-compatible insurance and agent documentation, not a citation booster today.

What belongs in llms.txt?

A one-line site description, then curated markdown links with short annotations: docs, key product pages, pricing, canonical explainers. Curation is the point — it's a map, not a sitemap dump.

Related terms