mcp-gui
Built MCP for multiple questions within one request (augument code tool)
claude mcp add --transport stdio oqwn-mcp-gui-server npx -y mcp-gui-server \ --env OPENROUTER_MODEL="MODEL YOUR SELECTED" \ --env OPENROUTER_API_KEY="YOUR API KEY" \ --env OPENROUTER_BASE_URL="https://openrouter.ai/api/v1"
How to use
The MCP GUI Server provides a modern graphical interface for collecting user input and orchestrating interactive prompts within MCP-enabled clients. It exposes an interactive web UI that supports image uploads, persistent terminal-like command execution, and real-time feedback workflows. The server integrates with multiple LLM providers through an abstraction layer, enabling prompt engineering enhancements and multi-provider routing. After enabling the GUI tool in your MCP client (via the gui-input tool), you can launch prompts that request structured user input, media uploads, or terminal commands, all within a secure session that can auto-expire and clean up when complete.
How to install
Prerequisites:
- Node.js 18+ (recommended)
- pnpm 8+ (recommended) or npm/yarn as an alternative
Option A: Run with NPX (no install required)
- Ensure you have Node.js installed
- Run: npx mcp-gui-server
Option B: Install from source (development or customization)
- Clone the repository: git clone https://github.com/oqwn/mcp-gui-server.git
- Navigate to the project directory: cd mcp-gui-server
- Install dependencies (pnpm recommended): pnpm install
- Build the project: pnpm run build
- Start the server: pnpm start
Prerequisites for development:
- pnpm installed globally (npm install -g pnpm)
- Optional: Docker or other deployment options if you prefer containerization
Additional notes
Tips and common considerations:
- If you encounter port 3501 in use, the server will auto-find an available port (e.g., 3517). Check console logs for the selected port.
- Environment variables under OPENROUTER_ are used to configure the OpenRouter integration. Replace placeholders with your API key, base URL, and model as needed.
- The NPX approach is ideal for testing; for production or custom deployments, install from source and run pnpm start to serve in your environment.
- The GUI supports image uploads (JPG, PNG, GIF, WebP) up to 10MB per image, with base64 embedding in MCP responses for multimodal workflows.
- If you run into NPX issues, consider clearing the npm cache or installing the package globally, then invoking mcp-gui-server directly.
- The server includes built-in session management, persistent terminal sessions, and hot-reloading during development.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud