Homebox MCP Server

Homebox MCP Server

Enables AI assistants to query and explore Homebox home inventory data in real-time, with tools for searching items, listing locations and labels, and retrieving item details.

Category
访问服务器

README

Homebox MCP Server

A Model Context Protocol (MCP) server that connects to your Homebox home inventory system, allowing AI assistants to query and explore your home inventory data in real-time.

What is This?

This MCP server acts as a bridge between AI assistants (like Claude) and your Homebox home inventory database. It allows your AI assistant to:

  • Search for items in your inventory
  • Get detailed information about specific items
  • List all locations where you store things
  • List all labels/categories you use
  • Find all items in a specific location
  • Find all items with a specific label

Deployment Options

Choose the integration method that works best for your use case:

For Open WebUI + Ollama Users

For Claude Desktop Users (MCP Protocol)

Which should I choose?

Prerequisites (Local Installation)

Note: If you're using Docker/QNAP, see DOCKER.md instead.

Before you begin, you'll need:

  1. Node.js - Version 18 or higher (Download here)
  2. A running Homebox instance - You should have Homebox already installed and accessible
  3. Homebox login credentials - Your email and password for Homebox

Step-by-Step Setup Instructions (Local Installation)

Step 1: Install Node.js

  1. Go to nodejs.org
  2. Download the LTS (Long Term Support) version
  3. Run the installer and follow the prompts
  4. Open a terminal/command prompt and verify installation:
    node --version
    npm --version
    
    Both commands should show version numbers.

Step 2: Download This MCP Server

If you haven't already:

git clone <your-repository-url>
cd Homebox-mcp

Step 3: Install Dependencies

In the Homebox-mcp folder, run:

npm install

This will download all the required libraries.

Step 4: Configure Your Homebox Connection

  1. Copy the example configuration file:

    cp config.json.example config.json
    
  2. Open config.json in a text editor and fill in your details:

    {
      "homeboxUrl": "http://localhost:7745",
      "email": "your-email@example.com",
      "password": "your-password"
    }
    

    Replace:

    • homeboxUrl: The URL where your Homebox is running (default is http://localhost:7745)
    • email: Your Homebox login email
    • password: Your Homebox login password
  3. Save the file

Step 5: Build the Server

Compile the TypeScript code:

npm run build

Step 6: Test the Connection

Before running the full server, test that it can connect to Homebox:

npm test

This will verify your configuration and show you if everything is working. If successful, you'll see green checkmarks and a summary of your inventory.

Alternatively, you can test the full server:

npm start

If successful, you should see:

Starting Homebox MCP Server...
Successfully authenticated with Homebox
Homebox MCP Server running on stdio

Press Ctrl+C to stop the server. You're ready to configure your AI assistant!

Configuring with Claude Desktop

To use this MCP server with Claude Desktop:

  1. Open your Claude Desktop configuration file:

    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
    • Linux: ~/.config/Claude/claude_desktop_config.json
  2. Add the Homebox MCP server to the configuration:

    {
      "mcpServers": {
        "homebox": {
          "command": "node",
          "args": ["/absolute/path/to/Homebox-mcp/dist/index.js"]
        }
      }
    }
    

    Replace /absolute/path/to/Homebox-mcp with the actual full path to your Homebox-mcp folder.

    Example for macOS/Linux:

    {
      "mcpServers": {
        "homebox": {
          "command": "node",
          "args": ["/home/username/Homebox-mcp/dist/index.js"]
        }
      }
    }
    

    Example for Windows:

    {
      "mcpServers": {
        "homebox": {
          "command": "node",
          "args": ["C:\\Users\\YourName\\Homebox-mcp\\dist\\index.js"]
        }
      }
    }
    
  3. Save the file and restart Claude Desktop

Using the MCP Server

Once configured, you can ask Claude to interact with your Homebox inventory. Here are some example queries:

Search for Items

"Can you search my Homebox inventory for 'screwdriver'?"

List All Locations

"Show me all the locations in my Homebox inventory"

Find Items in a Location

"What items do I have in the garage?" (first get the location ID using list_locations, then use get_items_by_location)

List All Labels

"What labels/categories do I use in Homebox?"

Get Item Details

"Can you get the full details for item ID 123?"

Find Items by Label

"Show me all items labeled as 'Electronics'"

For more detailed examples and use cases, see EXAMPLES.md.

Available Tools

The MCP server provides these tools:

  1. search_items - Search for items by name or description
  2. get_item - Get complete details about a specific item
  3. list_locations - List all storage locations
  4. get_location - Get details about a specific location
  5. list_labels - List all labels/categories
  6. get_label - Get details about a specific label
  7. get_items_by_location - Get all items in a location
  8. get_items_by_label - Get all items with a label

Troubleshooting

"Authentication failed"

  • Check that your config.json has the correct email and password
  • Verify that your Homebox instance is running
  • Make sure the homeboxUrl is correct

"Cannot find module"

  • Run npm install again
  • Make sure you ran npm run build

"ECONNREFUSED" or "Network Error"

  • Check that Homebox is running
  • Verify the URL in config.json is correct
  • If using Docker, make sure the Homebox container is running

Claude Desktop doesn't show the tools

  • Check that the path in claude_desktop_config.json is absolute (full path)
  • Restart Claude Desktop after changing the configuration
  • Check Claude Desktop logs for errors

Security Notes

  • Your config.json file contains your Homebox password - keep it secure
  • The .gitignore file is configured to prevent config.json from being committed to git
  • Never share your config.json file publicly

How It Works

MCP Server Approach (Claude Desktop)

This MCP server:

  1. Connects to your Homebox instance via its REST API
  2. Authenticates using your credentials
  3. Provides tools that Claude can use to query your inventory
  4. Returns data in a format that Claude can understand and present to you

The server doesn't directly access the SQLite database. Instead, it uses Homebox's official API, which is safer and more reliable for real-time access.

Open WebUI Functions Approach (Ollama)

The Open WebUI integration:

  1. Installs as a native function in Open WebUI
  2. Calls Homebox REST API directly from within Open WebUI
  3. Works with any function-calling capable Ollama model (Llama 3.1, Mistral, Qwen, etc.)
  4. No separate server process needed
  5. Perfect for Docker deployments on QNAP

See OPEN-WEBUI.md for complete setup instructions.

Development

If you want to modify the server:

  1. Edit files in the src/ directory
  2. Rebuild with npm run build
  3. Restart the server

License

MIT

Support

If you encounter issues:

  1. Check the Troubleshooting section above
  2. Verify your Homebox instance is working properly
  3. Check the Homebox documentation at homebox.software

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