2.4. Docs MCP Server

The Medusa documentation provides a remote Model Context Protocol (MCP) server that allows you to find information from the Medusa documentation right in your IDEs or AI tools, such as Cursor.

Medusa hosts a Streamable HTTP MCP server at https://docs.medusajs.com/mcp that you can add to AI agents supporting MCP server connections.

Authentication#

The Docs MCP server is available for Cloud users only. It requires authenticating with your Cloud account to use it.

The Docs MCP server supports two authentication methods:

1. OAuth: This is the recommended authentication method. It allows you to authenticate with your Cloud account without generating a token manually. When you connect to the Docs MCP server, you will be prompted to log into your Cloud account in a browser window to authenticate.
2. Personal Access Key: Use your Cloud personal access key as a Bearer token in the Authorization header when connecting to the Docs MCP server. This method is useful if you face issues with OAuth or prefer to use a token for authentication.

Connect to the Docs MCP Server#

Claude Code#

To connect to the Docs MCP server in Claude Code, run the following command in your terminal:

Then, start Claude Code and authenticate with your Cloud account:

Code
1claude2/mcp # for oauth

If you're connecting with OAuth, select the medusa MCP server and choose "Authenticate". This will open a browser window where you can log into your Cloud account to authenticate.

Codex#

To connect to the Docs MCP server in Codex, use one of the following methods:

If you're connecting with OAuth, run the following command in your terminal to authenticate with your Cloud account:

Terminal
❯codex mcp login medusa

This will open a browser window where you can log into your Cloud account to authenticate.

Cursor#

Add the following to your .cursor/mcp.json file or Cursor settings, as explained in the Cursor documentation:

If you're connecting with OAuth, open Settings -> Tools & MCPs and click "Connect" next to the medusa MCP server. This will open a browser window where you can log into your Cloud account to authenticate.

VS Code#

Click here to add the Docs MCP server to VSCode.

To manually connect to the Docs MCP server in VSCode, add the following to your .vscode/mcp.json file in your workspace:

If you're connecting with OAuth, once you start the MCP server connection, you will be prompted to authenticate with your Cloud account in a browser window. Log in to authenticate and start using the Docs MCP server in VSCode.

OpenCode#

Add the following to your OpenCode config file (for example, ~/.config/opencode/opencode.json):

If you're connecting with OAuth, run the following command in your terminal to authenticate with your Cloud account:

Terminal
❯opencode mcp auth medusa

This will open a browser window where you can log into your Cloud account to authenticate.

How to Use the Docs MCP Server#

After connecting to the Docs MCP server in your AI tool or IDE, you can start asking questions or requesting your AI assistant to build Medusa customizations. It will fetch the relevant information from the Medusa documentation and provide you with accurate answers, code snippets, and explanations.

For example, you can ask:

1. "Create a Product Review module for Medusa. Refer to the Medusa documentation for information."
2. "How to update Medusa to the latest version?"
3. "Explain the Medusa workflow system."
4. "Integrate Medusa with X provider. Refer to the Medusa documentation for information."

Docs MCP Server Tools#

Beyond answering questions from the documentation, the Docs MCP server provides tools that return curated, step-by-step implementation guides for common tasks. These guides tell your AI agent exactly which Medusa APIs and patterns to use, so it doesn't have to infer them.

The following chapters cover each tool, who it's useful for, and how to use it: