What is llms-full.txt?
Published:
If llms.txt is the table of contents, llms-full.txt is the whole book. The 2026 best practice is to publish both.
The two-file pattern
llms.txt gives a model a compact, linked overview — it has to fetch each link to read more. llms-full.txt embeds the complete content of those pages directly in a single file, so a model can ingest everything in one request with no extra fetches.
When to publish llms-full.txt
- You have documentation that benefits from being read in full (APIs, SDKs, guides).
- You want agents to work offline-ish, without crawling dozens of pages.
- You already maintain Markdown sources you can concatenate.
Tools like llms_txt2ctx can expand an llms.txt into a full version automatically.
Don't forget per-page .md versions
The spec also suggests offering a clean Markdown version of each page at the same URL with .md appended (and index.html.md for directory URLs). That gives models a parse-friendly copy of any single page.
Who does this
The llms.txt + llms-full.txt pairing is used by Anthropic, Vercel, and other documentation-heavy projects: the index for orientation, the full dump for deep ingestion.
Check yours
Our validator reports whether a sibling llms-full.txt exists when you check a domain. New to the format? Start with how to create an llms.txt file.