Do llms.txt and schema markup help you show up in AI answers?
By Ben Frampton · · 4 min read
Short answer: probably not much, on their own. There's no public evidence that adding an llms.txt file makes ChatGPT, Gemini, Perplexity or Google's AI features mention or cite you more, and Google has said its search features don't use it. Schema markup helps search engines understand your pages and can earn rich results, but it isn't a proven lever for being cited in AI answers. Both are cheap and harmless. Neither is a substitute for pages that answer buyer questions and rank.
What is llms.txt?
llms.txt is a proposed standard, published in September 2024 by Jeremy Howard of Answer.AI at llmstxt.org. It suggests putting a Markdown file at /llms.txt that summarises your site and links to the pages a language model should read, a bit like a sitemap written for an AI.
The idea is sensible: models have limited context, and a curated summary is easier to use than a whole website.
Do AI assistants use llms.txt?
As of October 2026, none of the major assistant providers has said that its answer engine reads llms.txt when deciding what to cite or recommend. Google has said publicly that it doesn't use it for Search, which includes AI Overviews and AI Mode. In server logs, requests for /llms.txt from the main AI search crawlers are rare.
Where llms.txt does get used is in developer tools: coding assistants and documentation tools that fetch a project's llms.txt to load its docs. If you publish developer documentation, it can be worth adding for that reason.
Our view: add one if it takes ten minutes and you like the idea. Don't expect it to change how often assistants recommend you.
What is schema markup?
Schema markup (structured data) is code, usually JSON-LD, that labels what's on a page using the schema.org vocabulary: this is an Organization, this is a Product with this price, this is an Article by this author published on this date.
Search engines use it to understand pages and to show rich results such as star ratings, prices and recipe cards.
Does schema markup help with AI search?
It helps indirectly, in three ways:
- Understanding. It can help search engines connect your pages, products and brand into a clear picture, which supports normal ranking.
- Consistency. Writing down your organisation's name, logo, social profiles and products in one structured place often exposes inconsistencies worth fixing.
- Eligibility for rich results in classic search, which can raise click-through.
What's missing is evidence that it directly makes AI answers mention or cite you. Assistants read your page's visible text to write their answers. Google's own guidance on AI features and your website says no special markup is needed to appear in AI Overviews or AI Mode, and that the normal SEO fundamentals apply.
One specific myth: adding FAQPage markup to get FAQ rich results. Since 2023 Google has shown FAQ rich results only for a small set of authoritative government and health sites. A visible FAQ section is still useful to readers and assistants. The markup alone does little.
What actually moves AI visibility?
From what can be measured across many answers, these matter far more:
- A page that directly answers the buyer's question. See how to write pages AI assistants can quote.
- Ranking for the searches the assistant runs. Retrieval rank is the best-supported predictor of citation. If your page isn't in the results the assistant reads, it can't be quoted.
- Mentions on the third-party sources assistants cite. Review sites, comparisons and community threads in your category. See the sources AI assistants cite.
- Crawler access. The search crawlers are allowed and can read your text. See which AI crawlers to allow in robots.txt.
That's why Notable never suggests llms.txt or schema markup as an action to improve AI visibility: we only suggest actions with evidence behind them, and we label how strong that evidence is.
A sensible checklist
| Item | Worth doing? | Why |
|---|---|---|
Organization schema on the homepage | Yes, once | Cheap, helps search engines identify your brand |
Product / Article schema where it fits | Yes | Supports rich results in classic search |
FAQPage schema for rich results | Rarely | Limited to a few site types since 2023 |
| llms.txt | Optional | Harmless; useful for developer docs |
| "AI instructions" pages written for bots | No | No evidence they work, and they can look manipulative |
Frequently asked questions
Is llms.txt the same as robots.txt?
No. robots.txt tells crawlers what they may fetch. llms.txt is a summary and reading list for language models. It grants or blocks nothing.
Will schema markup hurt my AI visibility?
No, as long as it's accurate and matches what's visible on the page. Markup that contradicts the page can be ignored or treated as spam.
Should I create a page of "instructions for AI"?
We don't recommend it. Assistants are built to resist instructions hidden in web pages, and a page written for bots rather than people adds no value for readers.
Could llms.txt matter more in the future?
It might, if providers adopt it. If that happens, it will be easy to add then. For now, time spent on useful pages and earned mentions is a better bet.