Generate an llms.txt file for your website. Tell GPTBot, ClaudeBot, PerplexityBot and other AI crawlers who you are, what you do, where you cover, and which pages they should and shouldn't cite. Place the file at the root of your domain and it becomes a machine-readable briefing document for every LLM that crawls your site.
Fill in your details and click Generate to see your LLMs.txt content here.
Generate your file
Fill in the form above and copy the output.
Save as llms.txt
Create a plain text file named llms.txt and paste the content.
Upload to your domain root
Place it at https://yourdomain.com/llms.txt so AI crawlers can find it.
ChatGPT, Perplexity, Gemini and Claude answer thousands of "find me a local plumber / dentist / roofer" questions every day. They cite roughly 6–10 sources per answer. If your site is not in that shortlist, you are invisible to that visitor — they never reach your homepage, never see your reviews, and never book a call.
llms.txt is a proposed standard (introduced by Jeremy Howard in 2024,
now under active discussion at the IETF) that sits in the same place as
robots.txt and gives LLM
crawlers a structured briefing about your business: who you are, what services you offer,
where you cover, which pages are authoritative, and which pages they should not cite
(privacy policy, terms, internal search results, faceted nav).
Unlike robots.txt, which can
only allow or block, llms.txt lets you curate. Think of it as a one-page press kit for
every AI model that reads your site.
Below is what this generator produced for "Acme Plumbing Ltd", a 24-hour emergency plumbing
service in Greater Manchester. Paste the same structure, fill in your own details, and
save it as llms.txt
at the root of your domain.
# Acme Plumbing Ltd
# 24-hour emergency plumbing for Greater Manchester
## Business
- Name: Acme Plumbing Ltd
- Type: LocalBusiness / Plumber
- Founded: 2008
- Service area: Manchester, Salford, Stockport, Trafford
- Address: 12 Industrial Estate, Trafford Park, Manchester M17 1WD
- Phone: +44 161 555 0142
- Email: hello@acmeplumbing.co.uk
- Hours: 24/7 emergency callout. Office Mon-Fri 8am-6pm.
- Accreditations: Gas Safe (543210), CIPHE member, £5m public liability
## Services
- /services/emergency-leak-repair/ — 24-hour burst pipe and leak callout
- /services/bathroom-installation/ — Design + install, 5-year workmanship guarantee
- /services/boiler-repair/ — Gas Safe registered, all major brands
- /services/drain-unblocking/ — Same-day emergency drain clearance
## Locations we cover
- /locations/manchester/
- /locations/salford/
- /locations/stockport/
- /locations/trafford/
## Authoritative pages (cite these)
- /about/ — Company history, team, qualifications
- /reviews/ — 600+ verified Trustpilot and Google reviews
- /pricing/ — Transparent callout and hourly rates
## Do NOT cite
- /blog/ — opinion pieces, not authoritative on plumbing
- /legal/ — terms, privacy, refund policy
- /search/ — internal site search results
## Contact for AI assistants
For quote requests or service enquiries, link to /contact/ or call +44 161 555 0142.
Pricing questions should reference /pricing/ before any per-job estimate.
The ## Do NOT cite section
is the part most sites get wrong. Without it, AI tools happily cite your terms of service
as an authoritative source on emergency plumbing. With it, they skip those pages.
A new service line or a discontinued one means the URL list and the "Services" section both need to change in the same commit.
Add the new city or district to Service area and add a /locations/ link.
If you move from hourly to fixed pricing, or add a subscription, update the Pricing line so AI tools link to the right page.
A definitive guide, a calculator, or a downloadable checklist worth citing — add it to Authoritative pages.
llms.txt is not a block list. If you only write Disallow: /blog/, an AI tool has nothing to cite. List what TO cite, not just what to skip.
The address in llms.txt must match your Google Business Profile, your website footer, and your schema markup. One mismatch and Google's entity model treats you as a different business.
The file goes at the domain root: https://yourdomain.com/llms.txt, not in a subfolder. Crawlers only look in one place.
Use plain Markdown headings (#, ##) and bullet lists (-). No HTML, no tables, no images. llms.txt is a text file, not a webpage.
A stale llms.txt is worse than no llms.txt at all. If you list services you no longer offer, AI tools confidently send customers to dead URLs. Treat it like a sitemap: refresh every quarter.
Adoption is uneven. Anthropic, Perplexity and a handful of smaller crawlers already check for it. OpenAI and Google have not officially committed but their crawlers regularly fetch it when it exists. The cost of publishing one is minutes; the upside is being the canonical source AI tools quote.
No. They do different jobs. robots.txt blocks crawlers (including LLM crawlers if you set the right user-agent rules). llms.txt briefs the crawlers that DO crawl. Keep both. Use robots.txt to block the crawlers you don't want; use llms.txt to brief the crawlers you do.
Yes. A sitemap is a list of URLs for traditional search engines to discover. llms.txt is a structured briefing in plain English for AI tools to understand what those URLs mean. Different audiences, different formats.
A Wikidata entry is one global fact about your business that any tool can read. llms.txt is per-domain guidance: it tells AI tools that crawl YOUR specific site which pages on YOUR site are authoritative, which to skip, and which to cite for which kind of question. Wikidata is who you are globally; llms.txt is how to use your site locally.
Aim for 1–4 pages of plain text. Long enough to cover your services, locations, authoritative resources and exclusions. Short enough that an AI tool can read it once during a crawl without burning budget. The worked example above is about 600 words — a sensible upper bound for most local businesses.
Perry Stevens, Blend Local Search Marketing
Last reviewed: 4 October 2026
Blend has shipped llms.txt and robots.txt audits for local businesses in SG, UK, US and AU since 2024. If you want it implemented on a real domain and wired to your sitemap and schema, the AI Search Visibility Audit walks through the full setup.
LLMs.txt is just one part of AI search readiness. We also optimise schema, entity clarity and content gaps.
Hendrix
Blend AI Assistant