Skip to content

LLMO Quickstart: Implement in 30 Minutes

You can make your site AI-discoverable in 30 minutes with three files. This guide covers the minimum viable LLMO implementation.

FilePurposeTime
robots.txtAllow AI crawlers to access your site5 min
llms.txtProvide AI with a structured summary of your site15 min
JSON-LD <script>Give AI structured data about your content10 min

Step 1: robots.txt for AI Crawlers (5 min)

Section titled “Step 1: robots.txt for AI Crawlers (5 min)”

Most sites already have a robots.txt. Add explicit Allow rules for AI crawlers:

User-agent: *
Allow: /
User-agent: GPTBot
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: Google-Extended
Allow: /
User-agent: PerplexityBot
Allow: /
Sitemap: https://yoursite.com/sitemap.xml

Why explicit Allow? Some hosting platforms and CDNs block AI crawlers by default. Explicit rules prevent accidental blocking.

CrawlerOperatorPurpose
GPTBotOpenAIChatGPT, web browsing
ClaudeBotAnthropicClaude web search
Google-ExtendedGoogleGemini, AI Overviews
PerplexityBotPerplexityPerplexity search
AmazonbotAmazonAlexa, product search
CCBotCommon CrawlTraining data collection

The llms.txt file (proposed by Jeremy Howard at llmstxt.org) gives AI systems a structured summary of your site.

Place this file at your site root: https://yoursite.com/llms.txt

# Your Site Name
> One-sentence description of what your site does.
## What We Do
A brief paragraph explaining your core offering, expertise, or purpose.
Use plain language. Avoid marketing jargon.
## Key Facts
- Founded: [year]
- Team: [size or key people]
- Location: [if relevant]
- Specialization: [your core expertise]
## Products / Services
- **Product A**: Brief description
- **Product B**: Brief description
## Links
- Website: https://yoursite.com
- Documentation: https://yoursite.com/docs
- GitHub: https://github.com/yourorg
- Contact: https://yoursite.com/contact
  1. Lead with facts, not marketing. “We build Android apps with AI automation” beats “We leverage cutting-edge synergies.”
  2. Include structured data. Tables, lists, and key-value pairs are easier for AI to parse than prose paragraphs.
  3. Keep it under 2,000 words. Concise summaries are more likely to be ingested fully.
  4. Update regularly. AI systems re-crawl periodically. Stale llms.txt means stale AI responses.

Add a JSON-LD script to your homepage <head>. This helps AI understand your entity type, relationships, and key attributes.

<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Your Company",
"url": "https://yoursite.com",
"description": "What your company does in one sentence.",
"founder": {
"@type": "Person",
"name": "Founder Name"
},
"sameAs": [
"https://github.com/yourorg",
"https://linkedin.com/company/yourorg",
"https://x.com/yourorg"
]
}
</script>
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "Your Article Title",
"author": {
"@type": "Person",
"name": "Author Name",
"url": "https://authorsite.com"
},
"datePublished": "2026-01-15",
"publisher": {
"@type": "Organization",
"name": "Your Company"
}
}
</script>
Content TypeSchemaPriority
HomepageOrganization or PersonHigh
Blog postsArticle or BlogPostingHigh
ProductsProductHigh
FAQ pagesFAQPageMedium
DocumentationTechArticleMedium
BooksBookMedium

After deploying, check:

  1. robots.txt: Visit https://yoursite.com/robots.txt and confirm AI crawlers are allowed
  2. llms.txt: Visit https://yoursite.com/llms.txt and verify the content is accurate
  3. JSON-LD: Use Google’s Rich Results Test or view page source to confirm the script tag is present
  4. AI test: Ask ChatGPT or Perplexity about your site/product and observe the response

This quickstart covers the Retrieval Signals and Structural Formatting components of the LLMO Framework. For the full framework: