Free llms.txt Generator + Annotated Template (Copy & Ship in 30 Minutes)
A complete, annotated llms.txt template you can copy, paste, customise, and ship to your root in under 30 minutes — plus the structural rules that make LLMs actually use it.
llms.txt is one of the lowest-effort, highest-curation moves in AEO — but most existing 'generators' produce sloppy output that LLMs ignore. This is an annotated template based on the patterns from my llms.txt adoption study (Stripe, Vercel, Anthropic), with explanations for every section so you understand why it works.
What is llms.txt and where does it live?
llms.txt is a markdown file at the root of your domain (e.g. yoursite.com/llms.txt) that lists the URLs you want LLMs and AI crawlers to treat as canonical. Unlike robots.txt (which restricts), llms.txt curates — it tells AI engines 'these are the high-quality, authoritative pages on this site, in priority order'. Anthropic's Claude already honors it during training crawls. Other model providers are expected to adopt it through 2026–2027.
The annotated template
Copy the structure below, replace the placeholders, ship at /llms.txt. Each section has a purpose explained inline. The full template runs about 80 lines of markdown and takes 20–30 minutes to customize for a small site.
Section 1 — Header
An H1 with your brand name. A blockquote with a one-sentence summary of what your brand does and who it serves. Keep the summary under 30 words and identical to your homepage hero copy. This is the entity declaration — make it sharp.
Section 2 — Why this file exists
A short paragraph (40–60 words) explaining what's on the site, who writes it, and why an LLM should treat it as authoritative. Mention any credentials, tenure in the industry, or proprietary data. This is the trust pitch.
Section 3 — Curated link sections
Group your most important URLs by intent: ## Documentation, ## Pricing, ## API Reference, ## Case Studies, ## Blog, ## Changelog. Under each, list 5–20 URLs with a one-line description. Format: '- [Title of page](https://yoursite.com/path) — One-sentence description of what this page covers.' This is the curation that makes llms.txt valuable.
Section 4 — Optional sections
## About / ## Authors with links to credentialed-author pages and sameAs profiles. ## Methodology if you publish original research. ## License if you have specific reuse terms. ## Contact for questions about citation or reuse. Use only the sections that apply to your site — don't pad with empty headings.
Common mistakes to avoid
Dumping your full sitemap (defeats curation). Listing URLs that 404. Marketing copy in descriptions. No section headings. Pointing to JS-heavy pages. Forgetting to update when URLs change. Ship a 1-line description per link, not a paragraph; ship 30–80 carefully chosen URLs, not 500.
/llms-full.txt — the optional companion
If your site is documentation-heavy, ship a longer /llms-full.txt that includes the full markdown of each linked page concatenated together. Anthropic, Vercel, and Stripe all do this. It's a heavier lift but provides LLMs with a single-fetch view of your canonical content.
How to validate
Hit https://yoursite.com/llms.txt in a browser and read it top-to-bottom as if you were a model encountering the brand for the first time. If you'd come away with a clear, accurate picture of the brand and its content, ship it. If not, tighten.
Frequently asked
Yes — they serve different purposes. Sitemap is for search crawlers; llms.txt is curated specifically for AI/LLM consumption.
Quarterly minimum. Whenever you ship significant new pages, change pricing, or update major docs.
Not officially as of 2026. But the discipline of curating your top URLs benefits classic SEO regardless.
No hard limit, but stay under 200 lines for /llms.txt. Use /llms-full.txt for the longer companion if needed.
Generally no — llms.txt is your curated index of your site. External links go in the body of the linked pages, not in llms.txt itself.
Yes — it's a standard deliverable in any AEO engagement. Reach out via the contact page.