Free LLMs.txt Checker — Validate Your /llms.txt File
Test your /llms.txt against the full llms.txt spec — H1 + blockquote, markdown link integrity, robots.txt cross-checks, AI bot accessibility (GPTBot, ClaudeBot, PerplexityBot, Google-Extended). AI-powered generator included.
First check can take ~20s to spin up — every one after starts instantly.
How It Works
Enter Your URL
Type your website address above. We locate /llms.txt automatically at the root of your domain.
We Analyze
Our checker runs every LLMs.txt-native check — H1 + blockquote, link integrity, robots.txt cross-checks, AI bot accessibility, file structure.
Generate or Fix
Get your score, a prioritized issue list, and a one-click AI generator if you don't have an llms.txt yet.
What is llms.txt?
llms.txt is a markdown file placed at the root of your website (example.com/llms.txt) that gives AI agents — ChatGPT, Claude, Perplexity, Gemini — a curated guide to your most valuable content. Where robots.txt says "don't go here," llms.txt says "here's what matters most."
The standard was proposed in September 2024 by Jeremy Howard of Answer.AI. It's now adopted by Anthropic, Vercel, Cloudflare, Mintlify, Hugging Face and thousands of smaller sites. The format is intentionally minimal — pure markdown — so any LLM can parse it without specialized libraries.
The companion file llms-full.txt takes a different approach: it inlines the entire content of your site so AI agents that need depth can grab everything in one request. Most large adopters ship both.
The llms.txt spec at a glance
# Your Site Name > One-sentence summary of the project — what it is, who it's for. ## Documentation - [Getting Started](https://example.com/docs/start): The five-minute setup - [API Reference](https://example.com/api): Full endpoint listing - [Examples](https://example.com/examples): Working code samples ## Guides - [Tutorial 1](https://example.com/tutorial-1) - [Tutorial 2](https://example.com/tutorial-2) ## Optional - [Changelog](https://example.com/changelog): Skippable for shorter context - [Press Coverage](https://example.com/press)
- Required: H1 with site/project name.
- Recommended: Blockquote summary directly after H1.
- Recommended: Each H2 section contains a list of markdown links.
- Optional special section: A heading literally called
## Optionalfor content AI agents can skip if context budget is tight. - Best practice: File must be served as
text/markdownortext/plain, nevertext/html. - Size: < 50 KB ideal, < 150 KB OK, ≥ 500 KB defeats the curation purpose — use llms-full.txt for bulk content.
Everything we check
Best practices
- ✓Curate, don't dump. An llms.txt with 30 high-quality links beats one with 300 noisy ones. AI agents allocate fixed context budget.
- ✓Group by intent. Use H2 sections that match how a user would ask: "Documentation," "Pricing," "Examples," "API Reference."
- ✓Use absolute URLs. Spec allows relative URLs, but absolute ones survive non-root deployments and partial fetches.
- ✓Add descriptions after a colon.
[Quickstart](url): 5-minute setupbeats a bare title — gives the AI signal about what each link is for. - ✓Don't block AI bots in robots.txt. Having an llms.txt while blocking GPTBot / ClaudeBot / Google-Extended is self-defeating.
- ✓Ship llms-full.txt too. For documentation-heavy sites, the full-content fallback unlocks better AI answers when context allows.
- ✓Update quarterly. Treat it like a sitemap — refresh when navigation changes or major content launches.
Frequently asked questions
Will AI agents actually read my llms.txt?
Major AI tools — ChatGPT browse mode, Claude, Perplexity, Gemini — check /llms.txt at inference time when answering site-specific queries. Adoption is growing fast. Even where adoption is partial, the cost is zero and the upside is direct control over how AI describes your product.
Should I block AI bots OR have an llms.txt?
They serve opposite purposes. If you publish an llms.txt, allow AI bots (GPTBot, ClaudeBot, PerplexityBot, Google-Extended) in robots.txt. If you actively want to keep AI agents out, block them in robots.txt and don't publish an llms.txt.
Can I autogenerate it?
Yes — our generator crawls your site, picks the top ~50 pages, then writes a spec-conformant llms.txt with curated sections. Edit before publishing. 50 credits per generation; refunded if generation fails.
Where exactly does it live?
HTTPS root: https://yourdomain.com/llms.txt — same convention as robots.txt and sitemap.xml. Subdirectory paths like /docs/llms.txt are not part of the spec.
What's the difference from sitemap.xml?
sitemap.xml is for traditional search-engine crawlers and lists every indexable URL. llms.txt is for AI agents at inference time and lists the curated subset you want them to load when answering questions about your site.
Related guides
Related tools
Common llms.txt mistakes to avoid
Linking to pages that no longer exist
An llms.txt is only useful if every link resolves. Dead or redirected URLs tell AI assistants your file is stale and untrustworthy. Validate every link against your live pages before you publish, and re-check whenever you restructure your site.
Skipping the H1 and summary blockquote
The spec opens with a single H1 — your site or product name — followed by a short blockquote that summarises what you do. AI models read that first to establish context. Omitting it makes the rest of your file harder to interpret correctly.
Blocking the same content in robots.txt
If your robots.txt disallows the very URLs you list in llms.txt, AI crawlers cannot fetch them. The two files should work together: robots.txt controls what is crawlable, llms.txt curates the best of what is allowed. Cross-check them so they do not contradict each other.
Dumping every URL instead of curating
llms.txt is a curated index, not a sitemap. List the pages that best explain your product, docs, and key resources — not every URL on the site. A focused file gives AI assistants a cleaner, more accurate picture of what matters.
Want a complete AI + SEO audit?
LLMs.txt is one of 20 checks we run. Get the full picture — robots.txt, sitemaps, meta tags, page speed, content quality, internal links, E-E-A-T, AI visibility — in one report.
Run Full Website Audit