What is llms-full.txt? Definition and when to use it
On this page
llms-full.txt is an informal convention for publishing a site's documentation as one Markdown file, so AI tools can load all of it in a single request.
How it works
It grew up next to llms.txtllms.txt is a proposed Markdown file at a website's root that gives AI tools a short summary of the site and links to the pages that matter most., but it is not part of that proposal. llms.txt is an index: a short Markdown file with a summary and links to the pages that matter. Version 2 of the proposal, revised in August 2026, says agents "view or search the llms.txt to find what they need, then follow the relevant links", and it asks sites to offer Markdown versions of pages at the same URL with .md added (llmstxt.org). Version 1 described a tool that expanded the file into one big context; version 2 dropped that tooling from the proposal (llmstxt.org changes).
llms-full.txt takes the other route: instead of links, it contains the text itself. Documentation platforms made it common by generating it automatically. Mintlify, for example, serves /llms-full.txt for every docs site and describes it as combining "your entire documentation site into a single file as context for AI tools". In that file, each page appears as its title, source URL, description and full Markdown content. Mintlify caps a generated llms.txt at 100,000 characters but says the limit does not apply to llms-full.txt (Mintlify).
Why it matters for your website
The file is useful when someone wants a whole manual inside an AI coding assistant or chat: one URL, pasted once, and the tool has everything. That makes most sense for developer documentation, API references and long product manuals.
It has costs:
- Size. A large docs site can produce a file of several megabytes, more than many tools fit in a context windowA context window is the maximum amount of text, measured in tokens, that a language model can take into account at once, including its instructions and reply., and every extra tokenTokens are the small units of text, often a word or part of a word, that language models read and write, and the unit in which AI usage is usually priced. costs time and money.
- Freshness. A hand-made file goes stale the moment a page changes. Generate it from the same source as the pages.
- Scope. Include only content that is already public. Drafts, internal notes and gated pages do not belong in a file anyone can download.
For a marketing site with a dozen key pages, a good llms.txt with links is enough. Our guide to llms.txt explains when each file is worth the effort.
Example
A shortened llms-full.txt for a small product's docs:
# Acme Analytics documentation
> Revenue dashboards for subscription businesses.
# Quick start
Source: https://acme.example/docs/start
Install the tracking script on every page, then connect Stripe.
# Billing FAQ
Source: https://acme.example/docs/billing-faq
Invoices are issued on the 1st of each month. Annual plans can be paid by bank transfer.A coding assistant given this one file can answer "How do I connect Stripe?" without fetching any other page.
Related terms
- llms.txt: the index file that llms-full.txt complements.
- Context window: how much text a model can read at once.
- Tokens: the units that file size is measured and billed in.
- AI crawler: the bots that may fetch such files.
- Chunking: the alternative of splitting content into small passages.
Sources
- The /llms.txt file (llmstxt.org) — accessed October 21, 2026
- Changes from v1 (llmstxt.org) — accessed October 21, 2026
- llms.txt (Mintlify documentation) — accessed October 21, 2026