vue-mcp-next
Enables real-time debugging and state manipulation of Vue.js applications through MCP protocol, integrating with Vue DevTools to access component trees, states, router info, and Pinia stores.
README
Vue MCP Next
📦 Installation
# Using npm
npm install vue-mcp-next
# Using yarn
yarn add vue-mcp-next
# Using pnpm
pnpm add vue-mcp-next
⚠️ Important Prerequisites
Before using Vue MCP Next, you must install and enable Vue DevTools!
Vue MCP Next relies on Vue DevTools to access and manipulate the runtime state of Vue applications. Please ensure:
- Install Vue DevTools browser extension
- DevTools is connected when the application is running
- Confirm DevTools can properly display component tree and state
Without Vue DevTools, Vue MCP Next will not work properly.
✨ Features
- 🚀 Standard MCP Protocol: Fully implemented based on official SDK
- 🏗️ Layered Architecture: Core layer focuses on protocol, Plugin layer adapts platforms
- 🔧 Multi-platform Support: Supports Vite, Webpack, Farm and other build tools
- ⚡ Runtime Operations: Real-time viewing and modification of Vue application state
- 🔍 Built-in Inspector: Integrated MCP Inspector debugging tool
- 📱 Vue DevTools Integration: Based on @vue/devtools-kit
🚀 Quick Start
Prerequisites
Before getting started, please ensure:
-
Install Vue DevTools browser extension
-
Ensure Vue DevTools can properly connect to your application
Vite Project Configuration
Add the plugin to your vite.config.ts:
import { defineConfig } from 'vite'
import vue from '@vitejs/plugin-vue'
import { vueMcpVitePlugin } from 'vue-mcp-next'
import DevTools from 'vite-plugin-vue-devtools'
export default defineConfig({
server: {
port: 5174 // Vite uses port 5174
},
plugins: [
vue(),
DevTools(), // Recommended: Enable Vue DevTools during development
vueMcpVitePlugin({
port: 8890, // MCP server port
inspector: {
enabled: true, // Enable MCP Inspector
autoStart: true, // Auto start
openBrowser: false, // Whether to automatically open browser
}
})
],
})
Start the Project
pnpm dev
After starting:
- Your Vue application will run at http://localhost:5174
- MCP server will run on port 8890
- MCP Inspector will start automatically (if enabled)
- Ensure the Vue DevTools extension in your browser can properly connect to your application
🔧 Configuration Options
VueMcpVitePlugin Options
interface VueMcpVitePluginOptions {
port?: number // MCP server port, default 8890
inspector?: {
enabled?: boolean // Enable Inspector, default true
autoStart?: boolean // Auto start, default false
openBrowser?: boolean // Auto open browser, default false
}
}
📋 Available Features
- Component Tree Viewing (
get-component-tree) - Component State Get/Edit (
get-component-state,edit-component-state) - Component Highlighting (
highlight-component) - Router Information (
get-router-info) - Pinia State Management (
get-pinia-tree,get-pinia-state)
💡 Use Cases
- Real-time Debugging: View and modify component state in Cursor
- State Validation: Verify component and application state meets expectations
- Interactive Testing: Simulate user interactions and state changes through AI commands
- Demo and Teaching: Showcase internal state and component structure of Vue applications
- Development Assistance: Quickly view router information, Pinia store state, etc.
A modern Vue.js Model Context Protocol implementation for Vue.js applications
Vue MCP Next is a modern MCP (Model Context Protocol) implementation that provides runtime state manipulation and control capabilities for Vue.js applications. This project is inspired by and built upon the excellent work of vite-plugin-vue-mcp, focusing on real-time manipulation of running Vue applications, supporting testing, debugging, demonstration scenarios, with layered architecture design, multi-build tool integration, and full compliance with official MCP protocol specifications.
🏗️ Project Structure
vue-mcp-next/
├── src/
│ ├── client/ # Client scripts
│ │ ├── client-script-manager.ts
│ │ └── devtools-client.ts
│ ├── plugins/ # Build tool plugins
│ │ ├── base/ # Base plugin functionality
│ │ └── vite/ # Vite plugin implementation
│ ├── server/ # MCP server
│ │ ├── devtools-layer.ts
│ │ ├── http-server.ts
│ │ └── server.ts
│ └── shared/ # Shared types and utilities
├── playground/ # Development test environment
└── dist/ # Build output
🛠️ Development
# Install dependencies
pnpm install
# Development mode
pnpm dev
# Start playground
pnpm dev:playground
# Build
pnpm build
# Test
pnpm test
# Type check
pnpm typecheck
📄 License
MIT License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。