Explain the site immediately
Use one H1 for the site name and a concise summary describing scope, audience, and essential context.
AI DISCOVERY
LLMs.txt Checker
Make AI-facing site guidance clear and structured.
Score AI-facing site guidance and resource links.
Start nowNo account required
Live
01Structure
02Links
03AI entry
Webmaster Technical SEO and AI Discoverability Guide
Learn how robots.txt, sitemap.xml, and llms.txt work together.
A domain checks /llms.txt at its root; a full file URL checks that exact location.
Scoring Notes
- Availability and text quality: HTTP 200 +20, plain-text Content-Type +5, valid UTF-8 +5.
- Core structure: one H1 on the first line +30; a one-line third-line summary (strongly recommended) +15.
- Detail content must not use headings +10; when H2 sections exist, they must contain link lists +15.
About llms.txt: provide curated entry points for AI inference
llms.txt is a Markdown-based proposal for giving language models and agents a concise site overview plus links to the most useful resources. It complements the broad URL inventory of a sitemap; it does not replace crawl control or search indexing mechanisms.
Under the format described by llmstxt.org, the only required section is an H1 naming the site or project. A following blockquote summary is strongly recommended, while optional H2 sections organize Markdown link lists. This checker scores whether the file is stable for programmatic and model consumption.
LLMs.txt Best Practices
Curate high-value links
Link core documentation, tools, policies, or entry pages, with short notes helping a model choose context.
Reserve Optional for secondary context
Place useful but skippable resources under ## Optional so smaller context windows can focus on essentials.
Maintain links with releases
Update llms.txt when core pages are published or moved so it does not recommend outdated documentation paths.
Frequently Asked Questions
Does llms.txt replace robots.txt?
No. Robots.txt communicates crawl rules, while llms.txt offers background and curated resources for model inference.
What structure is required in llms.txt?
According to the llmstxt.org proposal, the only required section is an H1 naming the site or project; a summary and H2 link sections are optional but useful.
Why place a summary immediately after the title?
A short summary gives the later resources context before a tool or model has to inspect every linked page.
Does llms.txt automatically improve search rankings?
There is no such guarantee. It is a proposal for clearer AI-facing context, not a guaranteed ranking signal.