What llms.txt adds
An llms.txt file summarises what your business is, which pages matter and which resources a model should read if it wants to understand you. It does not replace the sitemap or website content: it organises them for an AI context.
What WebMCP adds
WebMCP lets a page expose JavaScript tools with descriptions and schemas. For example, an agent could open the consultation questionnaire or read a service summary without guessing how the interface works.
How we apply it
At VisibleAI we add these pieces when they make sense: llms.txt for curated content, JSON-LD for entities and progressive WebMCP for important actions. Everything starts with a fast, accessible and clear website for humans.
- llms.txt map
- Schema per page
- Safe WebMCP tools
- Visible and verifiable content
How this guide was prepared
This guide combines official search and AI-platform documentation with technical criteria that small businesses can apply. We separate what a website controls—clarity, crawlability, performance, structure and conversion—from decisions made by third-party platforms.
Before prioritising a change, measure the baseline in Search Console and record where each enquiry comes from. That makes it possible to distinguish a useful improvement from a short-lived fluctuation without promising rankings or recommendations.
- Review indexing, queries and pages before publishing more content.
- Verify every business claim against a reliable source.
- Measure clicks, calls and messages, not only impressions.