The web's front desk is becoming an AI agent

Travelers used to find tours through ten blue links. Increasingly they ask an assistant — and the assistant reads your website on their behalf, decides whether it understands you, and either recommends you or silently moves on.

Generic AI-readiness scanners tell you whether crawlers can read your pages. That's necessary, but for a tour operator it's not the question. The question is whether an agent can complete the actual job: find the right tour, confirm the price and date, and book it. That's what the Bookability Score measures.

We weight six categories by their impact on that job. Tour Data & Schema carries the most weight (25%) because agents book from structured offers, not adjectives. Agent Bookability (20%) covers the new protocol layer — WebMCP, MCP, A2A, OpenAPI — plus the humble but critical test of whether "Book now" is a real link in your HTML.

Every check runs live against your site, every failed check ships with a concrete fix, and this site practices what it grades: scan webmcp.tours and hold us to our own standard.