What dependencies are required?
You need `@babel/preset-react` installed because the plugin traverses the AST to collect React component names.
@jazelly
基本信息
暂无概览
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"vite-react-mcp": {
"type": "http",
"url": "http://localhost:3000/sse"
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,直接连接服务器端点实时获取工具列表。
Vite React MCP is a Vite plugin that creates an MCP (Model Context Protocol) server to help LLMs understand your React app’s component structure, states, and performance. It is designed for developers using Vite and React who want to give AI assistants live insight into their application.
Install it as a dev dependency with pnpm install vite-react-mcp -D and also install @babel/preset-react. Add ReactMCP() to your Vite config’s plugins array. For MCP integration, configure your client: in Cursor, create ./cursor/mcp.json with a url property pointing to http://localhost:3000/sse. In Claude Desktop, a workaround script is needed because it expects command‑based execution. You can also access the tools directly via window.__VITE_REACT_MCP_TOOLS__ in the browser console.
You need `@babel/preset-react` installed because the plugin traverses the AST to collect React component names.
It uses `bippy` to inject a `__REACT_GLOBAL_DEVTOOLS_HOOK__` into the window, which provides access to React’s internal state without needing the DevTools extension.
The server exposes an HTTP endpoint via SSE at `http://localhost:3000/sse` (the same port as your Vite dev server).
Yes, but it requires a custom bridge script because Claude Desktop expects command‑based execution, while this MCP server is HTTP‑based. See [this discussion](https://github.com/orgs/modelcontextprotocol/discussions/16) for details.
No. By default it shows only self‑defined components. Pass `allComponent: true` to get the full tree.

Model Context Protocol Servers
This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,
Hindsight: Agent Memory That Learns

B2B competitive intelligence, live sales battlecards, 5-axis ICP prospect fit scoring, and tech stack teardowns over MCP and REST with cited public-web evidence.
A curated list of references for MLOps
评论