Is llms.txt an official standard?

It's a community convention (proposed by Jeremy Howard in 2024), not a ratified standard, and engines don't guarantee they read it. But it costs one small text file, several AI tools and crawlers do fetch it, and it gives every model a canonical self-description of your business to draw from instead of guessing.

How is llms.txt different from robots.txt?

robots.txt controls access — which crawlers may fetch which paths. llms.txt provides content — a curated summary of your business written for language models. They're complementary: robots.txt lets AI crawlers in, llms.txt tells them what matters.

Should llms.txt include pricing?

Yes, if your pricing is public. Concrete prices are exactly what buyers ask AI assistants about, and a machine-readable price list is the difference between an engine answering accurately and hallucinating. Generate the pricing block from your billing catalog so it never drifts from reality.

What is llms.txt and does your site need one?

By The SignalUp Team · Updated August 4, 2026

llms.txt is a plain-text (Markdown) file served at your site root — https://yoursite.com/llms.txt — that summarizes your business for AI language models: what you do, who it's for, what it costs, and where your important pages live. Think of it as a press kit written for machines.

Why it exists

AI assistants describe your business whether you help them or not. Left to infer from scattered page fragments, they get details wrong — old prices, discontinued products, a competitor's positioning. llms.txt gives every engine one canonical, current self-description to draw from. It's a community convention proposed in 2024, not a ratified standard, and engines don't guarantee they read it — but it costs one small file, several AI crawlers and tools do fetch it, and the downside is zero.

llms.txt vs robots.txt

robots.txt controls access: which crawlers may fetch which paths. llms.txt provides content: a curated summary written for language models. They're complementary — robots.txt lets AI crawlers in, llms.txt tells them what matters.

What to include

One-paragraph positioning — what you do, for whom, and your model (self-serve? sales-led?).
Capabilities — the concrete things you deliver, as short bullet claims.
Pricing — real numbers if public. Generate this block from your billing catalog so it never drifts from what you actually charge.
Key pages — canonical URLs for home, pricing, signup, legal.
Contact — a real email.

A real example

Ours is live at signalup.co/llms.txt. The skeleton:

# SignalUp

> One-paragraph positioning: what we do, for whom,
> how the ladder works, zero-meeting model.

## What SignalUp does
- Concrete capability claims, one per line.

## Pricing
(Generated from the billing catalog — never hand-typed.)
- Plan name: price, what it includes.

## Key pages
- Home, Pricing, Free scan, Legal — canonical URLs.

## Contact
- hello@signalup.co

Practical tips

Keep it under ~2,000 words — it's a summary, not a sitemap. Use absolute URLs. Regenerate it automatically when pricing changes; a stale llms.txt teaches every AI engine your old prices. And serve it as text/plain so nothing needs to parse HTML to read it.

llms.txt is one of the items on our AI visibility checklist — and one of the artifacts SignalUp ships automatically in Repair.

Is AI describing your business accurately? →
Free AI visibility report in minutes. No credit card, no sales call.