JetBrains IDE Integration MCP Server

The JetBrains MCP Proxy Server enables seamless integration between AI assistants and JetBrains IDEs like IntelliJ IDEA, PyCharm, WebStorm, and Android Studio. This integration allows AI models to access your project's code context, navigate files, and understand your development environment. With this tool, AI assistants can provide more relevant and contextual coding assistance by having direct access to your IDE's capabilities. The proxy server acts as a bridge between AI clients and the JetBrains IDE, enabling a more productive coding experience with AI assistance.

Overview

The JetBrains MCP Proxy Server creates a connection between AI assistants and JetBrains IDEs, allowing AI models to access your code context and provide more relevant assistance.

Installation

Step 1: Install the MCP Server Plugin in your JetBrains IDE

Open your JetBrains IDE (IntelliJ IDEA, PyCharm, WebStorm, etc.)
Go to Settings/Preferences → Plugins → Marketplace
Search for "MCP Server" and install the plugin
Restart your IDE when prompted

Step 2: Configure your AI client

For VS Code

You can use one-click installation or manual configuration:

One-click installation:

Use the VS Code extension marketplace to install

Manual configuration: Add the following to your VS Code User Settings (JSON):

{
  "mcp": {
    "servers": {
      "jetbrains": {
        "command": "npx",
        "args": ["-y", "@jetbrains/mcp-proxy"]
      }
    }
  }
}

Alternatively, you can add it to .vscode/mcp.json in your workspace.

For Claude Desktop

Add the following to your claude_desktop_config.json:

On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "jetbrains": {
      "command": "npx",
      "args": ["-y", "@jetbrains/mcp-proxy"]
    }
  }
}

After configuration, restart Claude Desktop with your JetBrains IDE already open.

Advanced Configuration

Connecting to a specific IDE instance

If you're running multiple IDEs with the MCP server plugin, you can connect to a specific one by adding:

"env": {
  "IDE_PORT": "<port of IDE's built-in webserver>"
}

Specifying a different host

By default, the proxy connects to IDE on 127.0.0.1, but you can specify a different address:

"env": {
  "HOST": "<host/address of IDE's built-in webserver>"
}

Enabling logging

To enable logging for troubleshooting:

"env": {
  "LOG_ENABLED": "true"
}

Troubleshooting

Node.js Version Requirements

If you encounter the error Cannot find module 'node:path', you need to upgrade to Node.js 18 or later. The MCP Proxy doesn't work with Node 16.

MacOS: Node.js Detection Issues with nvm

If you installed Node.js via nvm on macOS and the plugin can't detect it, create a symbolic link:

which npx &>/dev/null && sudo ln -sf "$(which npx)" /usr/local/bin/npx

External Client Connections

When connecting from external clients or Docker containers:

Enable "Can accept external connections" in IDE Settings → Build, Execution, Deployment → Debugger
Use your machine's LAN IP address and specify the IDE port:

"env": {
  "IDE_PORT": "YOUR_IDEA_PORT",
  "HOST": "YOUR_IDEA_LAN_IP"
}

Usage

Once configured, your AI assistant will be able to:

Access your project files and structure
Understand the context of your code
Provide more relevant coding assistance

Make sure your JetBrains IDE is running before starting your AI assistant client.