Need help with your APIs? I offer API discovery, governance & evangelism services. Explore services →
API Evangelist API Evangelist
Discovery
Learnings
Guidance
Toolbox
Alignment
API Evangelist LLC
Model Context Protocol (MCP) website screenshot

Model Context Protocol (MCP)

Model Context Protocol is an open, JSON-RPC 2.0 protocol from Anthropic that standardizes how AI applications (hosts) connect to external systems via servers that expose tools, resources, and prompts. MCP is positioned as "a USB-C port for AI applications" and has become the de-facto integration layer for Claude, ChatGPT, Cursor, Visual Studio Code, and a growing ecosystem of agent runtimes. This topic repo catalogs the canonical specification, the official multi-language SDKs, the reference server collection, community registries (Smithery, Pulse MCP, Naftiko Sandbox), and major MCP-aware clients, plus the vocabulary, JSON Schema, JSON-LD, and example payloads needed to reason about the protocol.

human only

Limited machine-readable signal and partial portal coverage — documentation a human can read, but little a machine or agent can consume without scraping.

Kin Score

API Evangelist profiles Model Context Protocol (MCP) the way a machine reads it — 66 machine-readable artifacts across 17 APIs, pulled from the provider's own public surface and indexed so a developer, an analyst, or an AI agent can evaluate it against every other provider on the network.

Every provider in the network is reduced to the same set of machine-readable artifacts — OpenAPI contracts, event specifications, GraphQL schemas, runnable collections, pricing and rate-limit signals, security posture, OAuth scopes, and the agent surfaces (MCP servers and skills) that let software drive the API on its own. We profile them because the interface is the part of a company you can actually inspect: it is a truer signal of what a provider does than any marketing page. From those artifacts we compute the Kin Score — Model Context Protocol (MCP) scores 31.8/100 (thin), with a separate agent-readiness read of 0/100 (human only). The full breakdown is below, followed by every artifact we hold — each card links through to its machine-readable definition on apis.io.

Kin Score

This is the API Evangelist rating — a single, repeatable read computed from the artifacts on this page. Green fill is points earned; the red track is points possible, so every bar shows earned-versus-possible at a glance.

Kin Score Kin Score How this is scored →
scored 2026-07-27 · rubric v0.5
Composite quality — 31.8/100 · thin
Contract Quality 8.0 / 25
Developer Ergonomics 3.9 / 20
Commercial Clarity 0.0 / 20
Operational Transparency 0.7 / 13
Governance 10.4 / 12
Discoverability 8.8 / 10
Agent readiness — 0/100 · human only
Machine-Readable Contract 0 / 18
Agentic Access Contract 0 / 15
MCP Server 0 / 12
Machine-Readable Auth 0 / 10
Idempotency 0 / 9
Stable Error Semantics 0 / 8
Request/Response Examples 0 / 7
Rate-Limit Signaling 0 / 7
Typed Event Surface 0 / 6
Agent Skills 0 / 5
Well-Known Catalog 0 / 4
Consent & Bot Identity 0 / 3

How we profile Model Context Protocol (MCP)

Each block below is one kind of artifact we hold for Model Context Protocol (MCP). For each we say what it is and why it earns a place in the profile, then list every one we've indexed — capped at two rows, scroll within the panel for the rest.

APIs 17

Each API is captured as its own OpenAPI definition — every operation, parameter, and response. This is the single most useful machine-readable description of what an API does, and it's what lets us score, lint, mock, and generate against it without asking the provider for anything.

Individual APIs this provider publishes, each with its own machine-readable definition.

MCP Specification

The authoritative protocol definition for the Model Context Protocol, maintained at modelcontextprotocol.io and in the modelcontextprotocol/ specification GitHub repository. The...

MCP TypeScript SDK

Official TypeScript SDK published as @modelcontextprotocol/sdk on npm. Dual-purpose library for building MCP servers and clients on Node.js, Bun, and Deno. Supports Standard Sch...

MCP Python SDK

Official Python SDK published as `mcp` on PyPI. Includes the FastMCP high-level framework with decorators for tools, resources, and prompts; low-level server primitives for prod...

MCP Java SDK

Official Java SDK for building MCP servers and clients on the JVM. Maintained in the modelcontextprotocol/java-sdk repository.

MCP Kotlin SDK

Official Kotlin SDK for MCP servers and clients targeting Kotlin and Android runtimes.

MCP C# SDK

Official C# SDK for .NET, maintained in collaboration with Microsoft.

MCP Swift SDK

Official Swift SDK for MCP servers and clients on Apple platforms.

MCP Rust SDK

Official Rust SDK for MCP servers and clients.

MCP Reference Servers

Canonical collection of reference MCP server implementations maintained by the project. The active servers (Everything, Fetch, Filesystem, Git, Memory, Sequential Thinking, Time...

MCP Inspector

Developer tool for testing and debugging MCP servers. React-based web UI plus Node.js proxy, launched with `npx @modelcontextprotocol/inspector` and accessible at http://localho...

Official MCP Registry

Community-driven, Anthropic-maintained registry that functions as an app store for MCP servers. Allows developers to publish server definitions and clients to discover them. Cur...

Smithery

Third-party MCP server registry and hosted runtime. Catalogs community-built MCP servers, generates installation snippets for clients like Claude Desktop and Cursor, and can hos...

Pulse MCP

Third-party MCP directory and news site. Maintains a sub-registry API that implements the Generic MCP Registry API specification with PulseMCP-specific extensions for enriched m...

Claude Desktop (MCP Host)

Anthropic's Claude Desktop application was the first MCP host. Connects to local MCP servers via stdio transport defined in claude_desktop_config .json and to remote servers via...

Cursor (MCP Host)

Cursor IDE supports MCP servers as a first-class extension mechanism, configured via mcp.json. Cursor is one of the most widely used MCP hosts for code-related workflows.

Visual Studio Code (MCP Host)

VS Code's GitHub Copilot Chat integrates MCP servers as agents and tools. Microsoft has standardized on MCP for Copilot's agent-mode extensibility.

ChatGPT (MCP Host)

OpenAI's ChatGPT supports MCP through the Apps SDK and developer tooling documented at developers.openai.com, making MCP a cross-vendor surface for connecting tools and data to ...

Scroll within the panel for all 17 ·

Features 12

The notable capabilities this provider advertises, captured as structured features so they can be searched and compared instead of read one landing page at a time.

Notable capabilities this provider offers.

JSON-RPC 2.0 Base Protocol

MCP defines request, response, and notification message shapes on top of JSON-RPC 2.0 with strict rules on IDs and structure.

Stateful Sessions

Clients and servers negotiate capabilities during the initialize handshake and maintain stateful connections for the lifetime of the session.

Tools

Servers expose callable tools with JSON Schema 2020-12 input schemas; clients invoke them via tools/call.

Resources

Servers expose URI-addressable context (files, database rows, API responses) via resources/list, resources/read, and subscription notifications.

Prompts

Servers expose templated prompts and workflows via prompts/list and prompts/get for users to invoke.

Sampling

Clients can offer sampling/createMessage so servers can ask the host LLM to run agentic, recursive inference under user consent.

Roots

Clients can advertise filesystem roots that bound a server's operating scope via roots/list.

Elicitation

Servers can request additional information from users mid-session via elicitation/create.

Multiple Transports

MCP defines stdio for local processes and Streamable HTTP (with an optional SSE legacy mode) for networked deployments.

OAuth 2.1 Authorization

HTTP-based transports follow an MCP-defined OAuth 2.1 authorization framework; stdio transports retrieve credentials from the environment.

Icons and Branding

Implementations, tools, prompts, and resources can publish icon metadata for richer UIs, with strict security constraints on icon URIs.

Reserved _meta Namespace

MCP reserves the _meta property with a structured prefix/name format for attaching additional metadata to interactions.

Scroll within the panel for all 12 ·

Semantic Vocabularies 1

JSON-LD contexts give the data shared meaning across APIs. We profile them because semantics are what let a machine reconcile 'customer' here with 'customer' somewhere else.

JSON-LD contexts and semantic vocabularies used across these APIs.

Mcp Context

34 classes · 12 properties

JSON-LD

Spectral Rules 1

Governance rulesets we run against this provider's specs — the automated checks behind parts of the score. Profiling them makes the quality bar explicit and re-runnable, not a matter of opinion.

Model Context Protocol (MCP) API Rules

6 rules · 4 warnings

SPECTRAL

JSON Schema 8

Standalone JSON Schema definitions describe the data models behind the API. We profile them so the shapes are validatable on their own — useful long after a single request is forgotten.

Standalone JSON Schema definitions for this provider's data models.

MCP tools/call Request and Result

0 properties

JSON SCHEMA

MCP Initialize Request and Result

0 properties

JSON SCHEMA

MCP JSON-RPC Message

0 properties

JSON SCHEMA

MCP Prompt

6 properties

JSON SCHEMA

MCP Resource

9 properties

JSON SCHEMA

MCP Root

3 properties

JSON SCHEMA

MCP Tool

8 properties

JSON SCHEMA

Scroll within the panel for all 8 ·

Examples 11

Real request and response payloads are what turn a spec from abstract into obvious — and they're one of the twelve things an agent needs to call an API correctly on the first try.

Example request and response payloads for these APIs.

Scroll within the panel for all 11 ·

Security Posture 2

Authentication, domain security, vulnerability disclosure, and trust-center signals — the evidence that a provider takes security seriously enough to document it. We profile it because you can't govern what you can't see.

Authentication, domain security, vulnerability disclosure, and trust-center signals.

Mcp Domain Security

TLSv1.3 · HSTS · DMARC

SECURITY

Mcp Vulnerability Disclosure

security.txt · contact published

SECURITY

Use Cases 5

What developers actually build with this provider — captured so the catalogue answers 'what is this for', not just 'what does this expose'.

What developers build with this provider.

IDE Augmentation

MCP servers extend coding assistants like Cursor, VS Code Copilot, and Claude Code with project-aware tools, repository search, and terminal/build access.

Knowledge Base Access

MCP resources surface documentation, wikis, and structured data stores to AI hosts under user consent.

Workflow Automation

MCP tools let agents trigger build pipelines, send notifications, create tickets, and orchestrate multi-step business workflows.

Enterprise Connectors

Vendors expose enterprise systems (CRMs, ITSM, data warehouses) as MCP servers so that any MCP-aware host can connect without bespoke integration code.

Local Computer Use

Local MCP servers expose filesystem, git, shell, browser, and desktop automation tools to assistants running on the same machine.

Integrations 5

Pre-built integrations with other platforms tell you where this provider already fits in a stack.

Pre-built integrations with other platforms and tools.

Anthropic Claude

Claude Desktop, Claude Code, and the Claude API are first-class MCP hosts and are the protocol's reference consumers.

OpenAI ChatGPT

ChatGPT supports MCP through the Apps SDK and OpenAI's developer docs.

Cursor

Cursor IDE consumes MCP servers via mcp.json configuration.

Visual Studio Code

VS Code's Copilot Chat treats MCP servers as agents and tools.

MCPJam

MCPJam is a community client implementation for testing MCP servers interactively.

Solutions 4

Packaged solutions the provider offers on top of the raw API surface.

Packaged solutions this provider offers.

Spec, SDKs, and Reference Servers

modelcontextprotocol.io publishes the spec, schema, official SDKs for TypeScript, Python, Java, Kotlin, C#, Swift, and Rust, plus a set of reference servers and an Inspector deb...

Official Registry

The official MCP Registry is a central index of MCP servers with GitHub OAuth/OIDC and DNS-based ownership verification.

Third-Party Registries

Smithery and Pulse MCP provide alternative discovery surfaces and, in Smithery's case, hosted execution of community servers.

Vendor Hosts

Claude, ChatGPT, Cursor, VS Code Copilot, Continue, Cline, Zed, Windsurf, and many other assistants act as MCP hosts.

Resources

Every other property we hold for Model Context Protocol (MCP) — documentation, portals, status pages, policies, and corporate surface — grouped by the job it does, following the integrator's arc from getting started to running in production.

Get Started 1

Portal, sign-up, and the first successful call

Documentation 1

Reference material describing how the API behaves

Design & Contract 2

Pagination, idempotency, versioning, errors, and events

Build 2

SDKs, sample code, and the tooling you integrate with

Access & Security 2

Authentication, authorization, and security posture

Company 1

The organization behind the API

← All providers · Data indexed from github.com/api-evangelist/mcp · machine-readable index on apis.io