Skip to main content

Product 2 · mystery shopper

Can an agent finish the job?

We crawl the origin, count real WebMCP tools (not a regex checkbox), and walk audit / book / checkout journeys. Missing tools are conversion risk. This is not an LLM ranking.

We do not sell LLM rankings. Nobody can measure them honestly. We sell what can be measured: structure, accessibility, and receipts to prove it.

Public tools on this site

12

Session tools (signed in)

6

Polyfill

On without Chrome flag

What we measure

  • Native document.modelContext vs polyfill vs none. Polyfill is not native Chrome.
  • Tool names and schemas, not a webmcp string match.
  • UI actuation (compare, highlight, navigate) vs JSON-only stubs.
  • Money paths: audit, book/consult, checkout/seal.