We sell machine readability. Here is ours.

llms.txt is the orientation file: what AgentFaro is, whom we serve, and where the canonical pages are, in the plain-text format the emerging llms.txt convention proposes. llms-full.txtexpands it with the full text of our answers and glossary, so a model can ingest the substance in one fetch.

Alongside those: sitemap.xml for crawlers, a human-readable index, and per-assistant files forChatGPT, Claude,Gemini, Copilot, andPerplexity, each stating our crawl policy for that vendor's agents and how we prefer to be cited.

Honesty requires a disclosure most sites skip: llms.txt is a real and growing convention, but the per-assistant files are not an established standard, and no assistant fetches them by default today. We publish them because they cost nothing, they are readable by anything that crawls, and if the convention arrives we are already standing there. Every file is generated from one source of truth, so they can never disagree with this website or with each other.

Most websites block AI crawlers, sometimes by choice, often by a template default nobody reviewed. For a business whose product is being findable by AI, that would be self-sabotage. Our robots.txt allows the major AI agents explicitly and by name.

The same check runs in every client audit, because an accidental block in robots.txt quietly removes an agent from the assistants they want to be named by.

Every page carries a schema.org graph: one Organization entity, consistent identifiers, and a typed node for the page itself, whether that is a service, a question with its answer, or a defined term. View source anywhere on this site and the graph is there to read.

The pages themselves are static HTML with a single stylesheet and no script dependencies, fully functional with JavaScript disabled. Fast for people, trivial for crawlers, nothing hidden behind rendering.