24 June 2026
All Topics
New feature
You can now build with AI using the commercetools AI plugin, a tooling package that gives your coding agent accurate knowledge of commercetools APIs and the proven implementation patterns you need to build storefronts and integrations.

With the plugin, you can go from an empty folder to a running storefront connected to your commercetools Project in a single working session. The agent handles SDK setup, authentication, data fetching, and checkout wiring, so you can focus on the parts that are unique to your implementation.

The plugin comes with the following Skills:
  • commercetools-platform: TypeScript SDK setup, authentication, the core data model (Products, Customers, Orders, Types, Channels, and Stores), GraphQL versus REST, query predicates, optimistic concurrency, and observability.
  • commercetools-storefront: a complete Next.js 16 Backend-for-Frontend (BFF) reference for B2C and B2B storefronts, including product listing, product detail, search, facets, cart, checkout, and customer authentication.
  • commercetools-checkout: integration with Checkout, for Payment Only, Complete Checkout, and Express Payment, and PSP connector setup for Stripe, Adyen, and Mollie.
  • commercetools-commerce-patterns: cross-cutting commerce logic covering pricing, discounts, shipping, tax, payments, B2B order flows, and bulk import.
  • commercetools-connect: building and shipping Connect applications, including API Extensions, Subscription and event handlers, scheduled jobs, and Merchant Center Custom Views.
The plugin also includes two MCP servers:
  • commercetools-knowledge: a public, read-only endpoint that handles live documentation search, schema lookup, and query validation with no API key required.
  • commerce-mcp: an optional server that lets the agent call your commercetools Project APIs directly using your credentials.
Three commands are also provided: /nextjs-setup-project scaffolds a new storefront, and /deploy-vercel and /deploy-netlify deploy it with automatic safety checks before anything goes live.
The plugin is tested with Claude Code, Cursor, VS Code with Copilot, and Codex. However, any agent that reads SKILL.md and works with the Model Context Protocol (MCP) can use it.