llms.txt
A proposed standard for websites to tell LLMs how to crawl and use their content.
llms.txt is a proposed convention (analogous to robots.txt) that lets site owners provide LLM-friendly context to AI crawlers. It typically lives at /llms.txt and contains a plain-text overview of the site plus links to the most important pages in markdown format. The structure usually includes a brief description of the site, section headings (## Directory, ## Content, ## About), and URL-title pairs under each section. Not yet universally adopted by AI crawlers as a formal standard, but increasingly common on documentation sites, SaaS products, and content-heavy platforms. For indie founders publishing content, adding an llms.txt is low-effort GEO hygiene that takes under an hour. Tools.cafe has its own llms.txt at https://tools.cafe/llms.txt. A practical template: write 2–3 sentences describing your product, list your 5–10 most important pages with descriptions, and add a brief note on what content you want AI tools to cite.
Related terms
GEO (Generative Engine Optimization)
SEO for AI-powered answers — Google AI Overviews, ChatGPT, Perplexity, Claude.
SEO (Search Engine Optimization)
The practice of getting a page to rank higher in Google's search results.
Schema markup
Structured JSON-LD data that tells search engines what a page is about.