What is Model Context Protocol?
The shift to agentic commerce
The way people shop is moving into AI. Discovery, comparison, checkout, and even post-purchase monitoring increasingly happen inside AI assistants rather than on a search results page or a brand's own site. This isn't a fringe behavior: 73% of consumers are already using AI in their shopping journey. And the money is following — Accenture estimates that by 2030, more than 30% of online commerce will run through AI agents, close to $3.1 trillion in transactions.
From scraping to brand-controlled protocols
Until now, AI assistants learned about products by scraping the open web – guessing at prices, availability, and specs from whatever pages they could read.
That era is ending.
Brands and the platforms behind these assistants are converging on agentic-commerce protocols: structured, brand-published feeds and endpoints that tell an AI exactly what you sell, what it costs, whether it's in stock, and how to complete a purchase. The difference is control – the brand, not a scraper, becomes the source of truth.
The catch is pace. Several of these protocols have emerged in just the last year, each backed by different platforms, and they ship new releases frequently. Staying compliant is an ongoing commitment, not a one-time integration. And each protocol takes a different approach.
What is MCP?
MCP – the Model Context Protocol – is an open standard for connecting AI applications to the external tools and data they need to be useful. The protocol's own analogy is that it works "like a USB-C port for AI applications": a single, standardized way to plug an assistant into your systems, instead of a one-off integration for every app and every model. With MCP, an AI assistant can call your tools, read your data, and run your workflows through one well-described interface.
It solves a real integration problem. Before there was a shared standard, every pairing of an AI app with a data source or tool needed its own bespoke connector. MCP replaces that convoluted mess with one protocol: build a server once, and any MCP-compatible client can use it.
How it works
MCP follows a client-server architecture. An AI application (the host) opens a connection via a client to one or more servers, each of which exposes some capability. Under the hood it's a JSON-RPC protocol with a clear lifecycle: the client and server complete an initialize handshake, negotiate a protocol version, and declare their capabilities before any work happens.
Servers offer three core primitives. Tools – executable actions an agent can invoke, resources – context data it can read, and prompts – reusable templates. Servers can run locally over the stdio transport or remotely over Streamable HTTP with standard authentication such as bearer tokens and OAuth. An agent discovers what a server offers at runtime – listing its tools, then calling them – so the surface stays dynamic rather than hard-coded.
Who's behind it
MCP was originally introduced by Anthropic and has since grown into an open-source project hosted by the Linux Foundation, developed openly by a broad community with official SDKs maintained across many languages. Adoption is wide: AI assistants including Claude and ChatGPT support it, as do developer tools such as Visual Studio Code and Cursor, alongside a large and growing ecosystem of servers.
Where it fits
MCP is the connectivity foundation of agentic commerce, not a commerce protocol itself. It's the transport and tool-calling layer that several commerce standards build on – for example, an ONX server is an MCP server exposing a standardized set of fulfillment tools. So getting MCP right is the groundwork: a clean, compliant MCP surface is what lets the commerce protocols layered on top actually reach an agent.
You can run a completely free check to see whether your MCP server measures up against the published MCP spec right now. And if you'd rather skip the technical implementation entirely, CorgiMaps connects to your existing data to expose a commerce-ready MCP server – and keeps it current as MCP and the protocols built on it evolve – with no changes to your fulfillment stack or operations.