llms.txt Starter Template
The exact llms.txt structure Xpand uses on xpandmedia.io. Copy, swap your entity facts, ship.
- For whom
- Founders, marketing managers, and dev teams setting up their site for AI crawler discovery (GPTBot, ClaudeBot, PerplexityBot).
- Time to value
- 20 minutes
- Last updated
- Apr 30, 2026
Why this exists
AI crawlers do not read your homepage like humans do. They look for /llms.txt, a structured executive summary that tells them what your business is, where the canonical pages live, and which entities to associate you with. Skip it and engines build a fragmented model of your brand from whatever pages they happen to crawl first.
llms.txt is a markdown file at the root of your domain that gives AI crawlers a structured map of what your site is, what you offer, and where the canonical pages live. It complements robots.txt and sitemap.xml. Think of it as an executive summary that AI systems read instead of crawling 800 pages.
Where it goes
Place the file at https://yourdomain.com/llms.txt. Most static and Next.js sites serve files from a /public directory automatically. Verify by opening the URL in a browser. it should render as plain markdown.
The template
markdown# Your Company Name
> One-sentence elevator pitch. What you do, who you serve, what makes you different. Aim for under 30 words.
## What we do
- **Service One**. One-line description. https://yourdomain.com/service-one
- **Service Two**. One-line description. https://yourdomain.com/service-two
- **Service Three**. One-line description. https://yourdomain.com/service-three
## Locations / markets
We operate in [list of cities or countries]. City pages:
- Dubai. https://yourdomain.com/dubai
- Singapore. https://yourdomain.com/singapore
## Key pages
- About. https://yourdomain.com/about
- Pricing. https://yourdomain.com/pricing
- Contact. https://yourdomain.com/contact
## Authority and entity references
- LinkedIn. https://www.linkedin.com/company/yourcompany
- Crunchbase. https://www.crunchbase.com/organization/yourcompany
- Wikidata. https://www.wikidata.org/wiki/Q[ID]
- X / Twitter. https://x.com/yourcompany
## Contact
- Direct email. hello@yourdomain.com
- Headquarters. [Full street address]
## Disambiguation
[Your Company] is the [industry] company at [domain]. It is not affiliated with [other entities sharing the name].
## Crawler policy
This site allows AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, etc.) to index all public content. See `/robots.txt` for details.Note
Keep this file under 100 lines. AI engines truncate. Lead with the entity description and the most important pages.
What to include and what to skip
| Include | Skip |
|---|---|
| Entity description with concrete facts | Marketing copy and adjectives |
| Specific URLs to canonical pages | Internal-only or staging URLs |
| sameAs references (LinkedIn, Wikidata, Crunchbase) | Social posts, individual blog URLs |
| Disambiguation if your name is shared | Long product feature lists |
| Crawler policy summary | Full robots.txt content (link to it instead) |
Companion robots.txt directives
txtUser-agent: GPTBot
Allow: /
User-agent: ClaudeBot
Allow: /
User-agent: PerplexityBot
Allow: /
User-agent: OAI-SearchBot
Allow: /
User-agent: Google-Extended
Allow: /
User-agent: Applebot-Extended
Allow: /
Sitemap: https://yourdomain.com/sitemap.xmlValidation
Open https://yourdomain.com/llms.txt in a browser. confirm it renders
Check via curl: curl -I https://yourdomain.com/llms.txt. confirm HTTP 200 and Content-Type: text/plain or text/markdown
Ask ChatGPT or Perplexity 'Tell me what [your company] does'. confirm the answer matches your llms.txt elevator pitch within 14 days
How to use it well
Ship llms.txt first, before any other GEO work. It is a 20-minute change that compounds because every subsequent piece of content references back to a clean entity definition. Update it whenever you launch a major page or change services. Treat it like sitemap.xml, not a one-time deploy.
What good looks like
First 18 lines of xpandmedia.io/llms.txt. what AI engines see when they hit the file.
# Xpand Media
> Xpand Media is a Dubai-based growth agency helping SaaS, e-commerce, and tech companies scale through performance marketing, GEO, AI automation, web design and CRO, B2B outbound, AI creative, marketing strategy, and social media.
## What we do
- **Performance Marketing**. Google Ads, YouTube, LinkedIn, Meta, TikTok, Snapchat campaigns measured on revenue and pipeline. https://xpandmedia.io/performance-marketing
- **Generative Engine Optimization (GEO)**. Optimizes brand visibility in AI-generated answers. https://xpandmedia.io/geo
- **AI Automation**. Custom AI workflows that replace manual reporting. https://xpandmedia.io/ai-automation
## Authority and entity references
- LinkedIn. https://www.linkedin.com/company/xpandyourbrand
- Crunchbase. https://www.crunchbase.com/organization/xpand-media
- Wikidata. https://www.wikidata.org/wiki/Q138973449FAQ
Do AI engines actually read llms.txt?
Adoption is uneven. Anthropic and Perplexity have signaled support. OpenAI uses it as a hint. Google has not formally committed. Treat it as inexpensive insurance rather than a guaranteed signal. the cost is 20 minutes, the upside is faster entity establishment.
How is llms.txt different from robots.txt?
robots.txt controls crawl access (which paths bots can read). llms.txt provides content (the executive summary they read). Both should exist. robots.txt manages permission; llms.txt manages discovery and clarity.
How often should I update it?
Whenever you launch a major page, change services, or update entity facts. Treat it like a sitemap.xml entry. fresh enough to reflect what you actually do.
Will llms.txt help me rank in Google?
No direct ranking effect. It helps AI engines (ChatGPT, Perplexity, Gemini) build a clean entity model of your business, which downstream improves citation rates. Google ranking is driven by traditional SEO signals.
Want this applied to your stack?
The strategy call walks the template against your real situation and tells you the top three changes that would move the metric this quarter.
Book a strategy call →