As someone leading both Adobe Commerce and Shopify practices at Valtech, I spend a good portion of my personal time experimenting with new tools, writing proof of concepts, and exploring how emerging technologies can improve the developer experience. One pattern I’ve observed is how platforms like Shopify and CommerceTools have been paving the way with Model Context Protocol (MCP) servers unlocking smoother AI-assisted development with tools like Cursor and Claude.
Yet, for Adobe Commerce developers, a clear gap existed.
There was no open-source MCP server tailored to the Adobe Commerce GraphQL APIs—no way to validate, autocomplete, or generate code in real time using AI based on the latest schema definitions.
That’s what motivated me to build the Adobe Commerce Dev MCP Server 💡
What Is It?
It’s an open-source MCP Server built specifically for Adobe Commerce. It helps devs code faster and smarter by enabling AI-powered tools to understand the context of Adobe Commerce’s GraphQL schema.
It’s now publicly listed on:
I’ve also published a quick YouTube demo showing how to set it up and use it.
Use Cases
🎯 Accelerate Adobe Commerce integrations
Code generation and autocompletion are aligned with the latest GraphQL APIs, directly inside AI-powered dev environments like Cursor.
🔍 Troubleshoot smarter
Identify deprecated fields or required inputs in mutations and queries, without digging through outdated docs.
✨ Create a better coding experience
Modernize how we work with Adobe Commerce. Let’s bring some of that “good vibe” experience from other ecosystems into Magento/Adobe Commerce.
What’s New in the Latest Version?
- Hosted on Vercel for instant usage, no local setup required
- Video tutorial included for easy onboarding
- Now listed in major MCP directories for AI dev tools
Let’s Build Together
This project began as a personal proof of concept, but with over 150 downloads and growing interest, I’m excited to see where the community can take it. If you’re working with Adobe Commerce, I’d love your feedback, suggestions—or even better—your contributions.
Special thanks to contributors for the early feedback that helped shape this release.