Free tool
llms.txt generator: create and validate the file
An llms.txt file is the curated entry file for AI agents: a Markdown file at the web root that bundles your name, your offering and your most important links. This tool builds it from your inputs following the llmstxt.org specification and checks existing files for format errors.
Your inputs
Sections and links
Group your most important pages. Links without a URL or label are skipped.
Your llms.txt
# Your file appears here once you enter the name …
Save the file as llms.txt at the root of your website so it resolves at yourdomain.com/llms.txt.
Validate an existing llms.txt
Paste an existing llms.txt. The check covers the H1, the summary blockquote, the order, sections, link format and size according to the llmstxt.org specification.
What is llms.txt?
llms.txt is a proposed convention: a Markdown file at the web root that states in a few lines who you are, what you offer and which pages matter. AI agents read it on demand, for instance when an assistant searches your website for specific information. Unlike the sitemap, which lists everything, llms.txt curates: it points agents at the pages that carry your claims. It is not a ranking signal; Google Search explicitly ignores it.
How to take the file live
- Generate the file and save it as llms.txt.
- Place it at the domain root (vincency.com/llms.txt).
- Serve it as UTF-8 (a BOM is allowed by the specification but unnecessary), Content-Type text/plain or text/markdown.
- Update it when content changes: a stale file is worse than none.
Common questions
Is llms.txt mandatory?
No. It is a proposal, not a W3C standard. None of the major providers commits to using the file for search or citations; Google states explicitly that Google Search ignores it. It is mainly used by AI agents and coding assistants that read documentation on purpose. With this generator it takes about ten minutes, by hand about an hour; its benefit for visibility in AI answers has not been shown so far.
What belongs in an llms.txt?
Only an H1 with the name is required. Recommended are a blockquote with one framing sentence and H2 sections with link lists in the format "- [Title](URL): note". Curated, not complete: we recommend the ten to thirty pages where your answers live.
Is llms.txt read by ChatGPT and Perplexity?
This has not been shown so far. OpenAI, Anthropic and Perplexity do not say in their crawler documentation that their crawlers use llms.txt. In an Ahrefs analysis of 137,000 domains, 97 percent of llms.txt files received not a single request in May 2026; the search bots of ChatGPT and Perplexity accounted for about 1 percent of requests. Among AI bots, coding agents and training crawlers fetch it most. Which pages a model finds is still decided by the search index and your content.
What is the difference from sitemap.xml?
The sitemap lists everything for search engines; llms.txt curates the essentials for AI agents. They complement each other: the sitemap says "all of this exists", llms.txt says "here are the answers".
Further reading
Sources
- llmstxt.org: The /llms.txt file, version 2 (modified 10 August 2026)
- Google Search Central: AI optimization guide (last updated 10 July 2026)
- Ahrefs: analysis of llms.txt requests on 137,000 domains (June 2026, updated August 2026)
- Chrome for Developers: Lighthouse llms.txt audit (experimental)
All sources retrieved on 15 October 2026.
Visibility is more than a file
llms.txt is the entry point, not the strategy. The seven layers above it and where your website actually stands are covered by the AI visibility check.
To the AI visibility check