Scorecard methodology

How we grade a WebMCP surface

01 How grading works

The scanner loads your website the same way an AI agent would and discovers every WebMCP tool it exposes. It then evaluates the implementation across two dimensions:

Technical implementation: Whether the WebMCP API is implemented correctly and reliably.
Tool quality: Are your tools actually useful for agents?

02 How we weight the final grade

The overall grade is a weighted blend of three scoring parameters.

Final grade weighting

Usability 60%

Coverage 20%

Quality 20%

Hover a parameter to see how it's scored

An agent reviews your tool surface, judging whether the tools are useful, clearly named, well described, and easy to call. Scored on a strict 1–5 scale.

5 — Exceptional: comprehensive coverage, precise schemas, unambiguous names. Genuinely rare.
3 — Decent: usable, but with real gaps. The typical good surface.
1 — Barely usable: an agent would struggle to call these reliably.

How much of your site is exposed to agents. Great implementations enable complete user journeys, not just the homepage.

1 page with tools → baseline
Each additional page raises coverage

The mechanical hygiene of each tool — the things that make a tool callable without guesswork. Averaged across every tool we find.

A real description (not a one-word stub)
A defined input schema the agent can fill
A clear, conventional snake_case name

03 Scores and grades

Each score maps to a letter grade.

Grade	Tier	Stars	What it means
A+	Exceptional	★★★★★	Best-in-class agent surface
A	Excellent	★★★★★	Strong, minor gaps only
A−	Very good	★★★★☆	Solid surface, a few fixes away
B+	Good	★★★★☆	Usable, real room to improve
B	Solid	★★★☆☆	Works, but gaps an agent will feel
B−	Needs work	★★★☆☆	Thin or loosely specified
C	Early	★★☆☆☆	WebMCP detected — the starting line

04 When there's no grade

A blank grade means one of these results:

api-absent: We didn't detect the WebMCP API. We looked for document.modelContext, navigator.modelContext, and declarative tool elements.
api-empty: The API was present, but no tools registered during the scan. If tools only appear after a user action, expose the discovery surface on page load.
blocked: The site blocks automated scans. That can hide your tools from agents and from this scorecard alike.
load-error: We couldn't load the page. Check that the URL is public, reachable, and not failing before tools register.

Want a better grade?

Book an implementation review →