Is llms-full.txt part of the official Answer.AI spec?

The original proposal describes tool-generated context files (llms-ctx.txt and llms-ctx-full.txt). The /llms-full.txt filename is a community convention popularized by documentation platforms like Mintlify — widely supported, but treat it as a convention rather than a ratified standard.

How big can llms-full.txt be?

There is no hard limit, but practical files run from 100 KB to a few MB. If your site is large, keep llms-full.txt to your most important pages — the generator caps automatic fetches at 50 pages and marks the result as a draft.

Definition

What is llms-full.txt?

Published 11 Aug 2026 · By Duy Nguyen

llms-full.txt is the companion file to llms.txt that inlines full page content as one markdown document. Where llms.txt is a short navigator of links, llms-full.txt contains the actual content of your pages, concatenated into a single self-contained file. An AI agent can read your entire documentation corpus in one HTTP request instead of following dozens of links — which is why llms-full.txt is favored for RAG pipelines, AI support bots and coding assistants with large context windows.

What does llms-full.txt look like?

The structure mirrors llms.txt but replaces links with content:

# Acme Docs — AI Reference

> **Audience**: AI search engines and AI agents.

## Documentation

### Getting started

Install the SDK with npm … (full page content)

### API reference

Every endpoint … (full page content)

Each page's markdown is included under its own heading, with heading levels bumped so page titles become H3s inside the combined document. Sections are separated by horizontal rules, and the file opens with an explicit audience blockquote.

Why do AI agents need a full-text file?

llms.txt tells an agent what exists; llms-full.txt lets it read everything without dozens of sequential fetches. For retrieval-augmented pipelines this is a major efficiency win: one request, full coverage, no crawling. It is the difference between a table of contents and the whole book.

How do you generate llms-full.txt?

The generator builds it from your sitemap: it derives titles and descriptions from your URLs, then renders a skeleton on the llms-full.txt tab in the editor. Skeleton pages carry placeholders you replace with real content, and the output is marked as a draft because it always deserves a human review before publishing.

When should you publish llms-full.txt?

If your site is documentation-heavy and you want AI support bots to answer accurately, publish it. If your content is volatile (pricing, product pages), prefer keeping the navigator only — llms-full.txt should be regenerated whenever content changes, or it can serve stale answers. For most marketing sites, llms.txt alone is the right starting point.