Guide
llms.txt: practical guide for AI citability
A machine-readable root file that explains brand, topics, and preferred pages to AI models and crawlers. · Aggiornato · Autore Engineering Factory · Chi siamo
Why it matters
llms.txt reduces entity ambiguity: it tells who you are, what you offer, and which URLs to cite. In Centropic, the presence and quality of the file feed AIO/GEO scores (Measured probe).
Answer engines and AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended) use public signals; a clear llms.txt is a citation anchor alongside Schema.org and permissive robots.
Useful sections
- Site / Summary / Key topics with canonical brand and domain
- Preferred citation (brand + expanded acronyms: AIO, GEO)
- Important pages with absolute https URLs
- Disambiguation (what you are not: GIS, generic All-in-One)
- Contact and owner / parent organization
How Centropic helps
Every analysis generates a draft llms.txt in the single HTML pack. Publish it at root (https://yourdomain/llms.txt) and re-analyze: the finding turns “available” once the probe finds it.
Live example: Centropic's own llms.txt is at centropic.ai/llms.txt, with preferred citation and a link to the methodology.
Quick checklist
- One single primary brand (avoid ghost names)
- Canonical https URLs, no tracking queries
- Align llms.txt with the JSON-LD Organization and the homepage
- Update the date whenever you change your offer or positioning
See also Schema AI and the methodology.