Guide · AI Crawlers
By Vlad Radu, Founder of Wolfdesk · July 9, 2026
Most websites still rely on robots.txt to communicate with crawlers, but robots.txt was built for search engines — not LLMs. llms.txt is an emerging standard that tells AI crawlers which pages matter, what they contain, and how they fit together. Think of it as a curated table of contents for ChatGPT, Perplexity, Gemini, and Claude.
llms.txt is a plain-text or markdown file served at the root of your domain. It lists the URLs you want LLM crawlers to prioritize, grouped by topic and summarized in a few lines. Unlike a sitemap, it is human-readable and designed to help models decide which pages are worth indexing.
The standard was proposed by Jeremy Howard and the Answer.AI team in 2024 and is now supported by a growing set of AI crawlers. It sits alongside robots.txt and sitemap.xml; it does not replace them.
Answer Engine Optimization depends on whether AI engines can extract and cite your content. If a crawler wastes its budget on obsolete pages or misses your most important guide, your citation rate drops. A well-written llms.txt:
Support is evolving. As of mid-2026, the following crawlers are known to consume or respect llms.txt:
Even where support is partial, publishing llms.txt is low-risk and positions your site correctly as the standard matures.
Create a markdown file at /llms.txt. Start with a short site description, then group links under headings. Keep summaries factual and concise — LLMs quote them.
# llms.txt — aeo-audit.app
> Free AEO audit tool that scores any URL on the technical, content, and authority signals that decide whether AI answer engines cite it.
## Product & guides
- [Home](https://aeo-audit.app/): Run a free AEO audit and see a citation-probability score, schema validation, and extractability report.
- [Blog](https://aeo-audit.app/blog): Field notes on Answer Engine Optimization and Generative Engine Optimization.
- [llms.txt guide](https://aeo-audit.app/blog/llms-txt-guide): How to write, deploy, and validate a llms.txt file for AI crawlers.
## AEO strategy
- [AEO strategy 2026](https://aeo-audit.app/blog/aeo-strategy-2026): The 2026 playbook for schema, llms.txt, E-E-A-T, and citation-rate measurement.
- [AEO vs GEO](https://aeo-audit.app/blog/aeo-vs-geo): Side-by-side comparison of Answer Engine Optimization and Generative Engine Optimization.
- [Check AI recommendations](https://aeo-audit.app/blog/how-to-check-if-chatgpt-recommends-your-website): How to check if ChatGPT, Perplexity, and Gemini are recommending your website.
## Legal
- [Privacy Policy](https://aeo-audit.app/privacy): GDPR data-processing and contact information.
- [Terms & Conditions](https://aeo-audit.app/terms): Service, price, and governing law.
- [Legal Notice](https://aeo-audit.app/legal): Operator details for Wolfdesk, Berlin.A useful llms.txt file is selective. It should not be a duplicate of your sitemap. Include the following:
We see the same errors across audits on aeo-audit.app:
First, confirm the file is reachable at /llms.txt with a 200 status and a text content type. Then check that it is linked from your robots.txt or site footer if you want broad discovery.
For a deeper check, run an audit. aeo-audit.app detects whether your site has a llms.txt file, previews its contents, and flags missing or malformed entries as part of the overall AEO score. It also validates schema, extractability, and crawler access in one pass.
No. robots.txt tells crawlers where they can or cannot go; llms.txt tells LLM crawlers which pages are worth reading and how they relate to each other. Use both. robots.txt handles permissions; llms.txt handles curation and context.
PerplexityBot, Claude's crawler, and several open-source LLM crawlers already consume /llms.txt. OpenAI and Google have signaled support but do not rely on it as a primary signal yet. The file is a forward-looking AEO investment that pays off as adoption grows.
Serve it at the root of your domain as /llms.txt, over HTTPS, with a text/plain or text/markdown content type. Some crawlers also accept /llms.txt.md or /.well-known/llms.txt, but the root path is the canonical location and the safest default.
Yes. llms.txt does not affect traditional search rankings. It only influences how LLM crawlers select and summarize content. You can edit it freely to prioritize your most citeable, up-to-date pages.
Next step
Run a free audit — llms.txt detection, schema validation, extractability score, and citation-probability rating.
Run free audit →