llms.txt Studio
What it does
Builds the site's llms.txt, the file proposed at llmstxt.org that tells language models which pages matter. It reads the site's sitemap (a sitemap index is followed one level), drops repeated URLs and groups the pages by the first part of their path: the home page under "Main", then one section per path such as "Blog" or "Products". It can also read each page's title and meta description, or rewrite the descriptions with AI.
It gives the same file as the free llms.txt Generator for the same sitemap.
When to use it
- When the brand wants to control which content LLMs see first.
- On very large sites, to anchor a set of priority pages (include or exclude paths, cap the number of URLs).
- For e-commerce sites with a long catalogue: the sections show the shape of the catalogue without forcing the crawler to go deep.
Required inputs
store_id(required): per-store metering, 3 runs/month.sitemap_url(optional): the sitemap to read; without it the site's sitemap is found on its own.max_entries(optional, default 100, at most 500),output_format(full,compact,sections-only).
Example output
{
"sitemap_url": "https://acme.it/sitemap.xml",
"content": "# Acme\n\n## Main\n\n- [acme.it](https://acme.it/)\n\n## Products\n\n- ...",
"entries_count": 100,
"truncated": true
}