How to Connect MCP Servers to JetBrains IDEs

JetBrains AI Assistant Now Supports MCP

JetBrains AI Assistant added MCP server support across all JetBrains IDEs: IntelliJ IDEA, PyCharm, WebStorm, GoLand, PhpStorm, Rider, and CLion. Your AI Assistant can now call external tools through MCP servers during conversations.

This means your IDE's AI can query databases, check dashboards, create tickets, and interact with any API you expose through an MCP server.

Prerequisites

• JetBrains IDE (2025.1+) with AI Assistant plugin
• AI Assistant subscription (included in JetBrains AI or All Products Pack)
• An MCP server endpoint (SSE or stdio)
• A DataFaucet account (free tier works) or any MCP server URL

Configuration

JetBrains reads MCP server config from your project's .jb-mcp.json or global settings.

Project-level config

Create .jb-mcp.json in your project root:

{
  "servers": [
    {
      "name": "your-app",
      "transport": "sse",
      "url": "https://datafaucet.dev/api/mcp/YOUR_SERVER_ID/sse"
    }
  ]
}

Global config (all projects)

Go to Settings > Tools > AI Assistant > MCP Servers and add:

• Name: your-app
• Transport: SSE
• URL: https://datafaucet.dev/api/mcp/YOUR_SERVER_ID/sse

Stdio transport (local servers)

{
  "servers": [
    {
      "name": "postgres",
      "transport": "stdio",
      "command": "npx",
      "args": ["@modelcontextprotocol/server-postgres", "postgresql://localhost/mydb"]
    }
  ]
}

Testing the Connection

1. Open AI Assistant chat (Alt+Enter or via tool window)
2. Ask something that requires your connected tool:

- "What tickets are assigned to me in Linear?"

- "Show me today's revenue from Stripe"

- "What's the status of my latest deployment?"

1. AI Assistant discovers tools and calls them inline

Multiple Servers

Add as many servers as needed. AI Assistant picks the right tool based on context:

{
  "servers": [
    { "name": "stripe", "transport": "sse", "url": "https://datafaucet.dev/api/mcp/STRIPE_ID/sse" },
    { "name": "jira", "transport": "sse", "url": "https://datafaucet.dev/api/mcp/JIRA_ID/sse" },
    { "name": "postgres", "transport": "stdio", "command": "npx", "args": ["@modelcontextprotocol/server-postgres", "postgresql://localhost/mydb"] }
  ]
}

DataFaucet + JetBrains Workflow

1. Go to DataFaucet and enter any URL
2. Browse the app for 30 seconds (API calls get captured)
3. Deploy your MCP server (one click)
4. Copy the SSE URL into .jb-mcp.json or IDE settings
5. Ask AI Assistant questions about that app's data

Total setup: under 2 minutes.

Common Use Cases

Code review with context:

"Pull the Jira ticket for this PR and check if my implementation matches the acceptance criteria."

Database queries while coding:

"What columns does the users table have? Show me the last 5 entries."

Deployment monitoring:

"Did the last deploy succeed? Any errors in the last hour?"

Project management:

"Create a Linear ticket for this bug with the stack trace I just saw."

Which JetBrains IDEs Support MCP?

All IDEs with AI Assistant plugin (2025.1+):

Troubleshooting

Server not connecting: Check the URL is reachable. Try curl https://datafaucet.dev/api/mcp/YOUR_ID/sse from terminal.

Tools not showing: Restart the IDE after editing .jb-mcp.json. Config is read at startup.

AI Assistant not using tools: Make sure you have AI Assistant Pro (not the free tier which may limit tool calls).

Next Steps

• Create your first MCP server (free, 60 seconds)
• Connect MCP to Claude Code
• Connect MCP to ChatGPT
• Best MCP clients compared