llms.txt
A proposed standard file placed at a site's root that offers AI systems a curated, plain-text map of the site's most important content.
llms.txt is a proposed convention — introduced in 2024 — for a markdown file at the root of a domain that gives AI systems a clean, curated guide to a site’s key content, analogous to how robots.txt guides crawlers. The goal is to help models find and use the most relevant, authoritative pages.
Important caveat
llms.txt is an emerging, voluntary proposal. As of 2026, adoption by major AI providers is limited and inconsistent, and there is no guarantee a given model reads or honors it. Treat it as low-cost, forward-looking hygiene rather than a reliable ranking lever.
How to use it
- List your most important, canonical pages with short descriptions
- Keep it accurate and current
- Don’t rely on it as your primary visibility strategy
Practical takeaway
Add it because it’s cheap and may help, not because it’s proven. Your retrievable on-page content and authority still do the heavy lifting.
Related Terms
- Optimization Tactics
Content Chunking
Structuring content into self-contained, clearly-labeled sections so AI systems can extract and cite a specific passage without needing the whole page.
- Optimization Tactics
GPTBot
OpenAI's web crawler, which gathers content used to train and inform its models; site owners can allow or block it via robots.txt.
- Optimization Tactics
Schema Markup
Structured data added to a web page using a shared vocabulary (schema.org) that explicitly tells search engines and AI systems what the content means.