What is llms.txt? The complete 2026 guide
The emerging standard that helps AI models understand your website. Spec, examples, and whether it's worth implementing right now.
Laxman Shah
Citegrade
TL;DR: llms.txt is a Markdown file at yoursite.com/llms.txt that gives AI models a curated map of your most important pages. Proposed by Jeremy Howard in 2024, it's the emerging standard for helping LLMs understand your site. OpenAI, Anthropic, and Perplexity have signaled support. It takes 30 minutes to create, costs nothing, and may improve how AI models retrieve and cite your content.
What is llms.txt?
llms.txt is a plain-text Markdown file placed at the root of your website (yoursite.com/llms.txt) that provides AI models with a structured guide to your most important pages and resources.
Think of it as a curated table of contents — written specifically for AI models that need to quickly understand what your site is about, what content matters most, and where to find authoritative information. Unlike a sitemap (which lists every URL), llms.txt is opinionated. It tells the AI: “here are the 15-20 pages that best represent us.”
The standard was proposed by Jeremy Howard, co-founder of Answer.AI, in 2024 as a response to a real problem: AI models often struggle to navigate large websites because their context windows are limited. A 200,000-word documentation site won't fit into any model's context. llms.txt solves this by providing a curated, highly compressed map.
Why does llms.txt exist?
Three problems pushed the standard forward:
- Context windows are finite. Even modern frontier models with 200K-1M token context windows can't fit a full SaaS documentation site. AI models need help prioritizing what to read.
- HTML is noisy for LLMs. Navigation menus, ads, footers, and JavaScript pollute the actual content. Markdown is the cleanest format for LLM consumption.
- Sitemaps weren't designed for AI. They're great for search crawlers but don't signal which pages are most important or what each page contains.
The solution: a single, hand-curated Markdown file that AI agents can read in one request to understand your site's purpose, structure, and priority pages.
The llms.txt specification
The format is intentionally simple. A valid llms.txt file contains, in order:
- An H1 heading with your site or project name (required, only one allowed)
- A blockquote summary explaining what your site is about (recommended)
- Optional intro paragraphs with key context
- H2 sections grouping related links by category
- Markdown link lists with descriptions:
- [Page title](url): What this page covers
That's it. No XML, no JSON, no rigid schema. Just structured Markdown that an LLM can parse trivially.
A complete llms.txt example
Here's a real-world example for a SaaS product:
# Citegrade
> Citegrade helps SaaS companies and websites get cited by ChatGPT,
> Perplexity, Gemini, and Claude. It scans your pages, identifies why
> AI skips you, and generates one-click fixes.
## Product
- [How Citegrade works](https://citegrade.com/about): Audit pipeline, scoring framework, and what makes content citable
- [Pricing](https://citegrade.com/pricing): Free, Pro ($9.99/mo), and Scale ($49.99/mo) plans
- [Sample report](https://citegrade.com/sample-report): Full audit example with scores, issues, and rewrites
## Features
- [Citation Checker](https://citegrade.com/features/citation-checker): Check if ChatGPT and Perplexity cite your page
- [AI Rewrites](https://citegrade.com/features/ai-rewrites): One-click fixes for citation blockers
- [Readiness Scoring](https://citegrade.com/features/readiness-scoring): 6-dimension content analysis
## Guides
- [How to get cited by ChatGPT](https://citegrade.com/blog/how-to-get-cited-by-chatgpt): 8 changes that increase citation rate
- [Why Google rankings don't mean AI citations](https://citegrade.com/blog/why-your-page-ranks-but-never-gets-cited): Structural reasons LLMs skip top-ranking pages
- [GEO vs AEO vs SEO](https://citegrade.com/blog/geo-vs-aeo-vs-seo): Three optimization disciplines explained
## Optional
- [Changelog](https://citegrade.com/changelog): Recent product updates
- [FAQ](https://citegrade.com/faq): Common questions about pricing, plans, and the audit processNotice the structure: site name (H1), one-sentence description (blockquote), then categorized link sections. Each link has a brief description that tells the AI what the page contains. The “Optional” section is the convention for content that's helpful but not essential.
llms.txt vs llms-full.txt
The standard supports two file variants:
- llms.txt — The concise version. 15-30 priority links, designed to fit comfortably in any LLM context window.
- llms-full.txt — The comprehensive version. Includes the full content of every page in Markdown format, not just links. Larger file, but lets the AI ingest your entire site in one request.
For most sites, llms.txt is enough. llms-full.txt makes sense for documentation-heavy products (think Stripe docs, Vercel docs) where AI agents need to answer technical questions accurately without re-fetching pages.
How llms.txt differs from robots.txt
They're complementary, not competing. Here's how they compare:
| Aspect | robots.txt | llms.txt |
|---|---|---|
| Purpose | Restrict crawler access | Guide AI to important content |
| Format | Plain text directives | Markdown with structured links |
| Audience | All web crawlers | AI models specifically |
| Tone | “Don't go here” | “Start here” |
| Adoption | Universal (since 1994) | Emerging (since 2024) |
You should have both. robots.txt sets the rules. llms.txt provides the map.
Who actually reads llms.txt today?
Honest answer: adoption is uneven. As of mid-2026:
- Documentation platforms — Mintlify, GitBook, and several other doc platforms actively generate and consume llms.txt files.
- AI coding tools — Cursor and similar AI agent tools use llms.txt to navigate documentation when answering developer questions.
- RAG systems — Many retrieval-augmented systems use llms.txt as a starting point for site indexing.
- Major AI labs — OpenAI, Anthropic, and Perplexity have signaled interest in the format, though none treat it as a definitive ranking signal yet.
The honest take: llms.txt is in the early-adopter phase. It's like sitemaps in 2005 — clearly useful, growing in support, but not yet a guaranteed ranking factor. The cost to implement is so low (30 minutes) that the upside-to-effort ratio is excellent.
How to create an llms.txt file in 30 minutes
Here's a fast workflow:
- List your top 15-20 pages. What URLs best represent your business? Include your homepage, key product/feature pages, pricing, top 3-5 blog posts or guides, documentation, and an “about” page.
- Write your H1 and summary. One H1 with your brand or product name. One blockquote that explains what you do in 1-2 sentences. Be specific — “a tool for X” not “an AI-powered platform.”
- Group pages by section. Use H2 headings: Product, Features, Pricing, Guides, etc. Keep groups intuitive.
- Write descriptive link text. Each link should follow the format:
- [Page title](URL): One-sentence description of what the page covers.The description matters more than the title. - Save as llms.txt. Place the file at your domain root. For Next.js or static sites, this means putting it in your
public/directory. For WordPress, upload via FTP or a plugin. - Verify it's accessible. Visit
yoursite.com/llms.txtin your browser. If you see your file rendered as plain text, you're done.
Is llms.txt worth implementing in 2026?
For most websites, yes. Here's the case:
Pros: Cheap to create (30 minutes). Forces useful clarity about your most important pages. Future-proofs your site for AI agents that increasingly use the standard. Some emerging platforms (Mintlify, GitBook) already use it actively.
Cons: Not a ranking signal in any major LLM yet. Easy to create a file that AI models ignore if your content quality is poor.
The mistake people make is treating llms.txt as a magic wand. It isn't. If your pages are vague, hedged, and unspecific, no amount of llms.txt curation will get you cited. The file is a map — AI still has to find the destination worth visiting.
That's why llms.txt should be paired with actual citation-readiness work. Audit your pages for the structural and semantic qualities that make AI cite content. Then publish the llms.txt file pointing AI at your strongest pages.
Bottom line: llms.txt is the emerging standard for telling AI models which pages on your site matter most. It's low-cost, low-risk, and adoption is growing. Create one in 30 minutes. But don't expect it to fix bad content — pair it with the editorial changes that actually drive AI citation. Audit your top pages to find out which ones belong in your llms.txt file.
Frequently asked questions
- What is llms.txt?
- llms.txt is a plain Markdown file placed at the root of your website (yoursite.com/llms.txt) that gives AI models a curated map of your most important pages. Proposed by Jeremy Howard of Answer.AI in 2024, it works like robots.txt but for LLMs — instead of restricting crawlers, it guides them to your highest-value content.
- Do AI models actually read llms.txt?
- Adoption is still growing. Several documentation platforms (Mintlify, GitBook) and AI coding tools (Cursor) actively consume llms.txt files, and major AI labs have signaled interest in the format. Major LLMs don't yet treat it as a definitive ranking signal, but it costs nothing to add and may improve retrieval quality for sites that implement it well.
- Is llms.txt different from robots.txt?
- Yes. robots.txt tells crawlers what they can't access. llms.txt tells AI models what they should prioritize. They serve different purposes — most sites should use both. robots.txt sets the boundary, llms.txt provides a roadmap within that boundary.
- Where do I put the llms.txt file?
- It must be at the root of your domain — yoursite.com/llms.txt. You can also create a more comprehensive llms-full.txt with additional context. Both should be plain Markdown files, served with content-type text/plain or text/markdown.
- Is llms.txt worth implementing in 2026?
- It's low-risk and potentially high-reward. Adoption is growing, the file takes 30 minutes to create, and it forces you to think clearly about which pages best represent your business. Even if AI models don't read it heavily today, it's part of the emerging standard for AI-first content discovery.