In this blog post
What Shopify MCP is and what it exposes to AI agents — the shopper side (Storefront & Customer Accounts MCP + UCP/Global Catalog) and the builder side (Shopify AI Toolkit).
AI agents are starting to shop. A customer can now ask ChatGPT to find a product, check if it's in stock, and be guided straight through to checkout — starting right from the chat. The piece of plumbing that makes this possible on Shopify is MCP, and it moves the first contact with your store into the assistant. Here's what it is, and exactly what it exposes to AI agents.
This is a deep-dive companion to our overview, Shopify Agentic Commerce: AI as Shopper vs. AI as Builder. Start there for the full map; stay here for MCP in detail.
What is Shopify MCP?
MCP stands for Model Context Protocol — an open standard that lets AI applications connect to external systems through one consistent interface. The easiest analogy is a USB-C port: instead of a different cable for every device, you get one universal connector. MCP is that universal port between AI agents (ChatGPT, Claude, Gemini and others) and your store's data and actions.
What does Shopify MCP expose to AI agents?
At its core, Shopify's MCP exposes your store's live data and commerce actions:
- Product catalogue — products, variants, descriptions, availability.
- Cart state — what an agent has added, in real time.
- Order information — when using an AI agent, logged-in customers can also perform tasks such as reordering and tracking orders.
- Store policies — shipping, returns, and other policy content.
- Checkout — completing the purchase through your store's own checkout, opened in the agent's in-app browser (your existing payment methods and branding apply).
Since 2026 the catalogue, cart and checkout actions run through Shopify's UCP endpoints rather than the original Storefront MCP tools. The table in the next-but-one section shows which endpoint does what.
In practice, this is what allows a ChatGPT shopping agent to not just mention your products but actually guide a customer through a full purchase — check stock in real time, add to cart, and redirect to your store's checkout in a ChatGPT in-app browser, where all your existing payment methods and customizations apply.
The Ecosystem Architecture: How UCP and MCP Connect
To fully understand Shopify’s agentic strategy, it helps to separate the map from the plumbing.
- Universal Commerce Protocol (UCP): the open ecosystem standard co-developed with Google (and backed by Amazon, Meta, Microsoft and others). UCP is the map — the shared language for how AI agents discover and transact with merchants. On Shopify you don't self-host a manifest: your products are made discoverable automatically through Shopify's Global Catalog, which aggregates listings from across the ecosystem so agents can find them. Agents themselves authenticate using a UCP profile hosted at a well-known URL.
- Model Context Protocol (MCP): once an agent discovers products via UCP, MCP is the protocol binding it uses to act on your store's live data. UCP says, “Here is a store that sells shoes,” and MCP lets the agent say, “Give me size 10 in blue, check inventory, and build a cart.”
Beyond text: enter “MCP UI”
One of the biggest omissions in casual discussions about MCP is that large language models natively understand only text and JSON. Walls of text are terrible for browsing highly visual industries like fashion, home decor, or beauty.
This is where MCP UI comes in — support for passing interactive UI components, not just text, through MCP.
- How it works: instead of the assistant just typing out a product description, an MCP server can return native, interactive visual components to the AI interface.
- The user experience: the shopper sees dynamic image galleries, price tags, interactive size/variant selectors, and “Add to Cart” buttons rendered inside the conversation (in a sandboxed frame) within assistants like ChatGPT.
Why this matters: MCP UI keeps agentic commerce from feeling like a 1980s text-based adventure game — your high-conversion, polished visual branding stays intact, right inside the conversational AI window.
| Layer | Protocol / Technology | What it Does for the Merchant | Why it Matters |
|---|---|---|---|
| 1. Discovery | UCP (Universal Commerce Protocol) | Broadcasts your inventory to global search engines and AI assistants. | Replaces legacy web scraping with clean, high-intent brand discovery. |
| 2. Action & Data | MCP (Model Context Protocol) | Provides a secure endpoint on your store for AI to read stock and modify carts. | Eliminates the need to build custom, messy APIs for every single AI app. |
| 3. Presentation | MCP UI | Renders interactive size/colour selectors and checkout frames inside the chat. | Keeps your premium visual branding intact so shopping doesn't feel like a text game. |
Two sides of Shopify MCP: AI as shopper, AI as builder
For a merchant, the useful split isn't by server name. It's by who benefits: your customers' AI assistants, or your own team's coding agents. The two sides have different owners, different budgets and different things to get ready.
1. The shopper side: demand that arrives through AI assistants. Every Shopify store exposes these endpoints, with no app to install. This is how Shopify's documentation splits the work today:
| What the agent does | Endpoint | Tools (Shopify docs) |
|---|---|---|
| finds products | Catalog MCP, via UCP (/api/ucp/mcp) | search_catalog, lookup_catalog, get_product |
| builds a cart | Cart MCP, via UCP | create_cart, get_cart, update_cart, cancel_cart |
| completes the purchase | Checkout MCP, via UCP | authenticated or signed requests only |
| answers policy and FAQ questions | Storefront MCP (/api/mcp) | search_shop_policies_and_faqs |
| tracks orders, handles returns | Order MCP and Customer Accounts MCP | for logged-in customers |
The original Storefront MCP catalogue and cart tools have been removed, so agents built on them have to move to UCP (Shopify docs). Through UCP and the Catalog API, eligible products can also show up beyond your own site: in ChatGPT, Copilot and the Shop App.
Who owns it: e-commerce and merchandising, not developers. Agents read your product data, variants, stock, policy pages and FAQs. In our projects the work here is mostly data: complete attributes and variant names, policy pages that answer the questions shoppers actually ask (delivery times, returns, sizing), and stock that matches reality. A shopper asks an assistant what to buy, and the stores it suggests are the ones whose data a model can read without guessing.
2. The builder side: how fast your team ships. The Shopify AI Toolkit (generally available since the Spring '26 Edition) gives coding agents such as Claude Code, Cursor and Codex access to Shopify's docs, GraphQL schemas, code validation and store operations. It includes the Dev MCP server (@shopify/dev-mcp).
Who owns it: your development team or agency. A human still reviews and ships every change. We describe how we use it in our AI-assisted Hydrogen workflow.
The shopper side earns you AI demand; the builder side changes how quickly you can act on it. Plan them separately.
Why this matters now
This isn't a someday story. With the Spring '26 Edition (June 2026), Shopify opened agentic commerce to every developer — any builder can now reach products from millions of Shopify merchants across AI surfaces like ChatGPT, Copilot and the Shop App. The standard underneath, UCP, was co-developed with Google and is backed by Amazon, Meta, Microsoft, Salesforce, Stripe, Etsy, Target and Wayfair — industry infrastructure, not a Shopify-only experiment.
And structured data already wins: Shopify reports that AI searches powered by its Catalog convert at 2× the rate of searches using scraped web data. The brands that own the new shelf aren't the ones with the best classic SEO — they're the ones whose product data, FAQs and policies are clean and legible to agents. (Source: Shopify, Spring '26 Edition, June 2026.)
What AI agents can — and can't — do
Through MCP, agents can read and write your store data: query the catalogue, manage products, pull reports, manage carts and orders. You can even run parts of your store in natural language from an interface like Claude.
What they cannot do is touch Shopify's proprietary core. AI agents can read and write store data via MCP, but they can't read or modify Shopify's backend business logic — that lives on Shopify's servers. When you need truly custom backend behaviour, that's a development question, not an MCP one.
Do you need a headless store for MCP?
For AI shoppers — no. A Shopify store can be exposed to AI channels through MCP regardless of whether it's Liquid or headless (Hydrogen). Architecture is irrelevant for basic visibility in ChatGPT or Gemini.
Headless matters more for AI-assisted engineering — the developer workflow with Claude Code and Cursor works far better on a clean Hydrogen/TypeScript codebase than on Liquid. We've documented that workflow end to end.
MCP for B2B: connecting to your stack
B2B agentic commerce is where MCP gets powerful — and complex. It means connecting MCP to your operational systems: ERP for pricing and availability, PIM for product specs, WMS for inventory. The Shopify B2B + headless + MCP stack is feasible, but the value is in careful integration and middleware design.
Getting started (and doing it safely)
Exposing your store to autonomous agents isn't a switch you flip. A production-ready setup involves MCP configuration, access controls, and guardrail design — explicit limits on what an agent can read or change without human approval. This is where most of the real engineering work (and risk) lives.
Want to make your store available to AI shoppers — safely? Talk to our Agentic Commerce team.
FAQ
What is Shopify MCP?
Shopify's implementation of the Model Context Protocol — an open standard (think USB-C for AI) that lets AI agents connect to your store's data and commerce actions through one interface.
What does Shopify MCP expose to AI agents?
Your live product catalogue, cart state, order information, store policies, and checkout — enough for an agent to discover products and guide a shopper through to your store's checkout.
What are the Shopify MCP servers?
Two sides. Shopper side: catalogue, cart and checkout through UCP endpoints (Catalog, Cart and Checkout MCP), policies and FAQs through the Storefront MCP server, and orders and returns through Order MCP and Customer Accounts MCP. Builder side: the Shopify AI Toolkit, including the Dev MCP server, for developers.
Do I need a headless store to use Shopify MCP?
No for AI-shopper visibility. Headless mainly helps the AI-assisted development workflow, not basic agent access.
Can I manage my Shopify store from Claude via MCP?
Yes, with setup — products, collections, content and reports through Claude, once MCP, access controls and guardrails are configured.
