nomos MCP Bridge

nomos MCP Bridge

A local MCP proxy that connects AI clients to nomos system controllers, allowing registration and switching between them via natural language. Enables smart home control through MCP-compatible AI clients.

Category
访问服务器

README

<div align="center">

<img src="https://github.com/nomos-system/node-red-contrib-nomos/raw/master/nodes-base/icons/nomos.svg" alt="nomos system" width="120" />

nomos MCP Bridge

A local MCP proxy that connects AI clients to one or more nomos system controllers.

npm version License: MIT

</div>


Instead of configuring each controller individually in your AI client, the bridge lets you register multiple controllers and switch between them via natural language.

Works with any MCP-compatible AI client — Claude Desktop, Cursor, Windsurf, ChatGPT Desktop, and more.

How It Works

MCP Client ──stdio──► nomos-mcp-bridge ──Streamable HTTP──► nomos Controller A
                      (local proxy)    ──Streamable HTTP──► nomos Controller B
                                       ──Streamable HTTP──► nomos Controller C

The bridge runs as a local MCP server (via stdio) and connects to nomos controllers over the network using the MCP Streamable HTTP transport. All tools, resources, and prompts from the connected controller are dynamically proxied — the bridge stays lightweight and always exposes exactly the capabilities the controller supports.

Prerequisites

  • Node.js 18 or later
  • A nomos system controller with MCP enabled (Skills > MCP) and a configured token
  • Network access to the controller from your machine

MCP Client Configuration

The bridge works with any AI client that supports the Model Context Protocol. No manual installation needed — just add the config and your client will run the bridge automatically via npx.

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "nomos": {
      "command": "npx",
      "args": ["-y", "nomos-mcp-bridge"]
    }
  }
}

Cursor

Add to your Cursor MCP settings (.cursor/mcp.json):

{
  "mcpServers": {
    "nomos": {
      "command": "npx",
      "args": ["-y", "nomos-mcp-bridge"]
    }
  }
}

Windsurf

Add to your Windsurf MCP config (~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "nomos": {
      "command": "npx",
      "args": ["-y", "nomos-mcp-bridge"]
    }
  }
}

Claude Code (CLI)

claude mcp add nomos -- npx -y nomos-mcp-bridge

Other MCP Clients

Any client supporting stdio-based MCP servers can use the bridge. The command is:

npx -y nomos-mcp-bridge

Adding Controllers

Option 1: Via the Setup Web Page

Tell your AI assistant to open the setup page:

"Open the nomos setup page"

This opens a local web UI in your browser where you can enter the controller name, URL, and MCP token.

Option 2: Via AI Chat

Tell your AI assistant to add a controller:

"Add my nomos controller 'Wohnhaus' at 192.168.1.100 with token abc123"

The assistant will use the add_controller tool to register it.

Option 3: Manually

Edit ~/.config/nomos-mcp/controllers.json:

{
  "controllers": [
    {
      "id": "some-uuid",
      "name": "Wohnhaus",
      "url": "https://192.168.1.100/mcp",
      "token": "your-mcp-token"
    }
  ],
  "activeControllerId": "some-uuid"
}

Usage

Once controllers are registered, simply tell your AI assistant which one to use:

"Connect to controller Wohnhaus"

"Switch to Büro controller"

"Show me all my controllers"

After connecting, all nomos tools are available as if the client were directly connected to the controller. You can:

  • Control devices ("Turn off the living room lights")
  • Create scenes and automations
  • Check system status
  • Configure the smart home

Bridge Tools

The bridge provides these management tools:

Tool Description
list_controllers List all registered controllers
select_controller Connect to a controller by name
add_controller Register a new controller
remove_controller Remove a registered controller
open_setup Open the setup web page in the browser
connection_status Show current connection status

Configuration

Controller credentials are stored in ~/.config/nomos-mcp/controllers.json. The setup web server runs on http://localhost:18900 (auto-increments if the port is in use).

Security Notes

  • Tokens are stored in plain text in the config file. Ensure appropriate file permissions.
  • The setup web server only listens on 127.0.0.1 (localhost) and is not accessible from the network.
  • nomos controllers use HTTPS with self-signed certificates by default. Node.js may reject these — set NODE_TLS_REJECT_UNAUTHORIZED=0 in your MCP client config if needed:
{
  "mcpServers": {
    "nomos": {
      "command": "npx",
      "args": ["-y", "nomos-mcp-bridge"],
      "env": {
        "NODE_TLS_REJECT_UNAUTHORIZED": "0"
      }
    }
  }
}

Development

# Clone the repository
git clone https://github.com/nomos-system/nomos-mcp-bridge.git
cd nomos-mcp-bridge

# Install dependencies and build
npm install
npm run build

# Run
npm run start

# Or build and run in one step
npm run dev

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选