Ga naar inhoud

WHAT IS WEBMCP?

WebMCP, A2A, and MCP are the protocols that let AI agents interact with your website — not just read it. We audit exactly how ready you are.

WHAT WE AUDIT

Nine checkpoints covering the agent protocol layer — the interfaces AI agents use to discover, understand, and interact with your website's functionality.

WEBMCP DECLARATIVE API

HTML meta tags that expose your site's forms and actions as tool definitions for AI agents. The easiest way to make your website agent-interactive without building an API.

WEBMCP MANIFEST

A JSON manifest at /.well-known/webmcp that describes all available tools, their input schemas, and endpoints. The machine-readable directory for your site's agent capabilities.

A2A AGENT CARD

Google's Agent-to-Agent protocol. A JSON file at /.well-known/agent.json that describes your site's agent capabilities, skills, and supported interaction modes.

MCP DISCOVERY

Model Context Protocol server discovery via /.well-known/mcp.json. Lets AI agents like Claude find and connect to your MCP endpoint for tool-use interactions.

OPENAPI SPEC

An OpenAPI/Swagger specification that documents your API endpoints. AI agents use this to understand available operations, parameters, and response schemas.

AGENTS.JSON

A directory file listing all agents and protocols your site supports. Helps AI systems discover which interaction methods are available in one lookup.

FORM QUALITY

How well your HTML forms are structured for agent interaction — labels, field types, autocomplete attributes, and accessibility that lets agents fill forms correctly.

MANIFEST SCHEMA QUALITY

Validates that your WebMCP manifest follows the spec — proper tool definitions, valid JSON schemas, required fields, and type annotations that agents can rely on.

INTERACTIVE SURFACE COVERAGE

What percentage of your site's interactive elements (forms, buttons, actions) are exposed as agent-callable tools via WebMCP or similar protocols.

WHY AGENT PROTOCOLS MATTER

AI agents are evolving from reading websites to interacting with them. These protocols are the bridge.

PROTOCOL ADOPTION
<3

Less than 3% of websites implement any agent protocol

CHECKPOINTS
0

Covering WebMCP, A2A, MCP, OpenAPI, and agents.json

TYPICAL SCORE
0

Most websites score near zero on agent protocols

EARLY MOVER ADVANTAGE
0

Of websites haven't started — implementing now puts you ahead

WANT THE COMPLETE PICTURE?

WebMCP Audit focuses on agent protocols. For the full 5-category AI readiness audit covering discovery, structured data, semantics, protocols, and security — visit IsAgentReady.

Full AI Readiness Audit →

FREQUENTLY ASKED QUESTIONS

Common questions about WebMCP, agent protocols, and how we audit them.

WebMCP (Web Model Context Protocol) is a standard that lets websites expose their functionality as tools that AI agents can call. Using HTML meta tags or a JSON manifest at /.well-known/webmcp, you declare what actions your site supports — like searching, booking, or submitting forms — so agents can interact with your site programmatically.

A2A is Google's Agent-to-Agent protocol. An Agent Card at /.well-known/agent.json describes your site's capabilities, supported interaction modes, and skills in a machine-readable format. It lets AI agents discover what your site can do before attempting to interact with it.

MCP (Model Context Protocol) Discovery via /.well-known/mcp.json lets AI agents find your MCP server endpoint. MCP is Anthropic's protocol for tool-use interactions — agents connect to your MCP server to call tools, access resources, and complete tasks on behalf of users.

Agent protocols are brand new — WebMCP, A2A, and MCP Discovery all emerged in 2024-2025. Most web developers haven't heard of them yet. Less than 3% of websites implement any agent protocol. This means early adopters have a significant competitive advantage as AI agent usage grows.

We check 9 agent protocol checkpoints, each worth different points: WebMCP Declarative API (25 pts), WebMCP manifest (10 pts), A2A Agent Card (25 pts), MCP Discovery (10 pts), OpenAPI (20 pts), agents.json (10 pts), Form Quality (20 pts), Manifest Schema Quality (15 pts), and Interactive Surface Coverage (20 pts). Your score is the percentage of points earned.

WebMCP Audit focuses exclusively on agent protocols — the 9 checkpoints that determine whether AI agents can interact with your site. IsAgentReady is the full 5-category audit covering AI Content Discovery, AI Search Signals, Content & Semantics, Agent Protocols, and Security & Trust across 47 checkpoints.

EXPLORE MORE

Most websites score under 5 on agent protocols. Find out where you stand.

AUDIT NOW
TEST YOUR SITE

AUDIT NOW

Enter your URL and get a detailed WebMCP audit in under a minute.
RANKINGS
SEE WHO LEADS

RANKINGS

See how websites rank on agent protocol readiness.