Setup Guide
This guide provides configuration instructions for connecting your AI development tool to the Vaadin MCP server. The Vaadin MCP server gives AI assistants direct access to comprehensive Vaadin documentation, enabling accurate and context-aware responses.
Choose Your AI Tool
Select your AI development tool from the tabs above for specific configuration instructions:
-
Claude Code - Anthropic’s official CLI tool with native HTTP MCP support
-
Cursor - AI-powered code editor with native HTTP MCP support
-
Windsurf - Codeium’s AI-native IDE with built-in HTTP MCP support
-
Junie - JetBrains AI assistant (requires stdio-to-HTTP adapter)
-
GitHub Copilot - GitHub’s AI pair programmer in VS Code and JetBrains IDEs Agent mode
-
Codex - OpenAI’s code generation tool with native HTTP support
-
Gemini CLI - Google’s Gemini CLI with native HTTP support
-
Other Tools - Configuration for any MCP-compatible client
Transport Types
|
Important
|
The Vaadin MCP server uses HTTP transport (streamable-http). This is natively supported by Claude Code, Cursor, Windsurf, GitHub Copilot (VS Code and JetBrains IDEs), Codex, and Gemini CLI. For stdio-only tools like Junie, you can use @pyroprompts/mcp-stdio-to-streamable-http-adapter as a bridge between the two transport types. |
General Setup Steps
Most tools follow these general steps:
-
Locate your tool’s MCP configuration file (varies by tool)
-
Add the Vaadin MCP server configuration
-
Restart your tool to load the MCP server
-
Start asking questions about Vaadin development
See the specific instructions for your tool in the tabs above for exact configuration details.