llms.txt Generator
Turn your page list into a valid llms.txt — correct H1, blockquote summary, absolute URLs, and every bad line called out by number.
The llms.txt convention asks for one file at the site root that an AI crawler can read in a single request. The format is tiny and easy to get slightly wrong — H1 first, the summary as a blockquote right under it, absolute URLs, no empty sections. Paste your page list and take the file.
How to use
Fill the site root, the site name for the H1, and a one-line summary, then paste your entries into the third box, one per line, written as `path | title | note`. Start a new section with a `## Heading` line. Relative paths are resolved against the site root, absolute URLs are kept as they are, and the same URL twice is dropped with a warning that names the first line it appeared on. Titles longer than 60 characters and notes longer than 160 characters are clipped and reported instead of being silently shortened; a missing title falls back to the last path segment, title-cased. Only the first pipe-separated field is required, so `path` alone works, and `path | title` works too. Tick the robots box if you also want the Allow block that points the five big AI crawlers at the file — that block needs an absolute site root, and without one it is skipped with an explicit error rather than emitted half-formed. The output is the file itself, ready to save as llms.txt at the site root, with the problem list above it so you can fix the input and rebuild.
FAQ
What is llms.txt? A plain-text file at the site root that lists the pages worth an AI crawler's attention, in markdown with links, so a model can find the good pages without parsing your whole navigation. Why does the summary have to be a blockquote? That is the shape the convention specifies — a single `>` line directly under the H1 — and it is the sentence an agent reads to decide whether the rest of the file is worth loading. Why are my relative paths dropped? A relative entry needs a site root to resolve against; without one there is nothing to make it absolute, so those lines are dropped by number instead of guessed at. Is there a size limit? No hard limit in the convention, and the tool reports the byte count and the longest URL so you can judge readability yourself. Does it replace sitemap.xml? No — sitemap.xml tells search engines what exists, llms.txt tells AI readers what matters. Ship both. Should I block the AI crawlers instead? If you would rather not be read at all, use the robots.txt option on the sitemap builder to disallow them; this option does the opposite and explicitly allows the file.
Privacy: every calculation runs inside your browser. Your files and text never leave your device.