Docs

Docs

Junie

Learn how to configure JetBrains Junie AI assistant to access Vaadin documentation through the Model Context Protocol server

Junie is JetBrains AI assistant for IntelliJ-based IDEs. It uses stdio transport and requires an adapter to connect to HTTP-based MCP servers like the Vaadin MCP server.

Requirements:

  • Node.js 18 or later

  • Junie plugin for IntelliJ-based IDEs

Configuration

  1. Open the IDE settings: Ctrl+Alt+S on Windows and Linux, Cmd+, on macOS.

  2. Go to Tools › Junie › MCP Settings.

  3. Click Add and paste the following configuration into the mcp.json file that opens:

Source code
JSON
{
  "mcpServers": {
    "vaadin": {
      "command": "npx",
      "args": ["@pyroprompts/mcp-stdio-to-streamable-http-adapter"],
      "env": {
        "URI": "https://mcp.vaadin.com/docs",
        "MCP_NAME": "vaadin"
      }
    }
  }
}

The adapter bridges stdio transport to HTTP, allowing Junie to communicate with the Vaadin MCP server. URI is the server’s address, and MCP_NAME is the name the server appears under.

Note

Servers added through the MCP Settings page are saved to ~/.junie/mcp/mcp.json in your home directory and are available in every project. To limit the server to one project, save the configuration as .junie/mcp/mcp.json in that project’s root instead.

How the Adapter Works

The @pyroprompts/mcp-stdio-to-streamable-http-adapter package:

  1. Receives stdio commands from Junie

  2. Translates them to HTTP requests

  3. Forwards requests to the Vaadin MCP server

  4. Returns responses back through stdio

This allows stdio-only tools to work with HTTP-based MCP servers.

Verify the Setup

After configuration:

  1. Check the Status column of the server list in Tools › Junie › MCP Settings. Expand it to see the Vaadin server’s tools.

  2. Open the Junie assistant panel

  3. Ask a question about Vaadin development (e.g., "What are Vaadin components?")

  4. Junie queries the Vaadin MCP server through the adapter

Resources

Updated –