Why MCP matters
The Model Context Protocol (MCP) is emerging as one of the most exciting standards in AI-assisted development. It defines a structured way for AI agents and developer tools, like GitHub Copilot, Cursor, or other AI-powered IDEs, to connect with external data and services. Instead of “chatting” in isolation, an MCP-enabled AI can reach out to servers that expose tools, resources, and prompts, making it aware of your environment, APIs, and documentation. In practice, this means developers can stay in the flow while the AI fetches examples, schemas, or code references from live sources, not static knowledge.
Sitecore MCP Server
To help developers build faster on its composable stack, Sitecore has released an MCP Server for its documentation and APIs. This server exposes the official developer docs, including XM Cloud, Edge Delivery API, Headless SXA, and JSS, in a way that AI tools can understand. When connected, your coding assistant can query Sitecore’s documentation contextually:
“Show me how to call the Edge Delivery API in Next.js.”
“What fields are returned by a content item query?”
“Give me an example fetch request for the XM Cloud API.”
Under the hood, the Sitecore MCP Server communicates over HTTPS, responding with structured JSON that MCP-aware assistants can process directly. It doesn’t replace the documentation; it amplifies it. By embedding the right knowledge where you need it the most: right inside your editor.
Adding it to VS Code
In this step-by-step process, we will add the Sitecore MCP server:
- Open VS Code
- Open your workspace
- Press [Ctrl] + [Shift] + [P]
- Search for MCP
- Choose MCP:Add Server
- Choose HTTP (HTTP or Server-sent Events)
- Enter https://doc.sitecore.com/mcp
- Press [Enter]
- Name it whatever you like, a suggestion would be Sitecore Documentation
- Press [Enter]
- Choose if you want it to be installed in the Workspace or Global
- Verify that the Server has been added
- In order for it to work you need to add a header on line 6"headers": {"User-Agent": "sitecore-mcp-client/1.0"}
- The result should look like this:
- Happy codin'!
Closing thoughts
AI is moving from passive autocomplete to active collaboration, what we now call Agentic AI. By exposing its documentation through an MCP Server, Sitecore is giving frontend developers a head start: faster answers, richer context, and fewer context switches.
For those building headless experiences with XM Cloud and Vercel, this is more than convenience. It’s the start of an AI-enhanced developer workflow. The combination of Sitecore, MCP and Agentic AI marks the next chapter in how we build, learn, and ship experiences faster than ever.