IntelliGlow
A Model Context Protocol (MCP) server that allows AI assistants like Claude to control real smart bulbs via UDP network communication, featuring voice commands, AI reasoning, and direct hardware control.
README
💡 IntelliGlow - AI-Powered Smart Lighting
"Smart lighting, brilliantly simple"
IntelliGlow is a Model Context Protocol (MCP) server that allows AI assistants like Claude to control real smart bulbs via UDP network communication. This Python implementation features voice commands, AI reasoning, and direct hardware control.
🏗️ Architecture
Voice/AI ──> IntelliGlow MCP ──> UDP Network ──> Smart Bulb (192.168.1.45:4000)
The smart bulb system that actually thinks!
🌟 Features
🔴 Real Hardware Support
- UDP Network Communication: Direct communication with real smart bulbs
- Default Bulb Configuration: Connects to
192.168.1.45:4000by default - Network Discovery: Automatically find smart bulbs on your network
- Connection Management: Persistent connections with auto-reconnect
🎤 Voice Intelligence
- Natural Voice Commands: "Turn on lights", "Set brightness to 75", "Make it blue"
- AI-Powered Parsing: Understands context and natural language
- Text-to-Speech Feedback: Speaks responses back to you
- Smart Color Recognition: Recognizes color names and descriptive terms
🧠 AI Integration
- MCP Protocol: Works with Claude, GPT, and other AI models
- Context Understanding: AI can reason about lighting needs
- Workflow Integration: Bulbs become part of larger AI workflows
- Learning Capability: Can adapt to user patterns and preferences
🔧 Smart Bulb Control
- Power Control: Turn bulbs on/off via UDP commands
- Brightness Control: Adjust brightness levels (0-100%)
- Color Control: Full RGB control with hex color codes (#FF0000)
- Status Monitoring: Get real-time bulb status
- Ping/Connectivity: Test network connectivity to bulbs
🌐 Network Features
- Multi-bulb Support: Connect to multiple bulbs simultaneously
- Discovery: Scan network for available smart bulbs
- Environment Configuration: Set bulb IP/port via environment variables
🚀 Quick Start
Installation
-
Install IntelliGlow:
# Core system pip install -e . # With voice capabilities pip install -e .[voice] -
Configure your bulb (optional):
export BULB_IP=192.168.1.45 # Your bulb's IP export BULB_PORT=4000 # Your bulb's port
Running IntelliGlow
# 1. MCP server only (for AI integration)
mcp-server-smartbulb
# 2. Voice interface only
mcp-server-smartbulb-voice
# 3. Complete IntelliGlow system (voice + AI + MCP)
python voice_enabled_server.py
Testing Network Connectivity
# Test UDP communication with your real bulb
python test_network_bulbs.py
🔧 AI Integration (Claude Desktop)
Add this to your Claude Desktop claude_desktop_config.json:
{
"mcpServers": {
"intelliglow": {
"command": "python",
"args": ["-m", "mcp_server_smartbulb.network_server"],
"cwd": "/path/to/your/IntelliGlow",
"env": {
"BULB_IP": "192.168.1.45",
"BULB_PORT": "4000"
}
}
}
}
🛠️ Available Commands
🎤 Voice Commands
- "Turn on the lights" - Power control
- "Set brightness to 75 percent" - Brightness with smart parsing
- "Make it blue" - Color recognition
- "How are the lights?" - Status inquiry
- "Find smart bulbs" - Network discovery
🤖 MCP Tools (for AI)
discover_bulbs()- Find smart bulbs on the networkconnect_to_bulb(ip, port)- Connect to a specific bulbturn_on_bulb(ip, port)- Turn on a bulb via UDPturn_off_bulb(ip, port)- Turn off a bulb via UDPset_bulb_brightness(brightness, ip, port)- Set brightness (0-100)set_bulb_color(color, ip, port)- Set color using hex codesget_bulb_status(ip, port)- Get current bulb statusping_bulb(ip, port)- Test connectivity to a bulb
📡 Network Configuration
Default Bulb Setup
IntelliGlow connects to 192.168.1.45:4000 by default. You can override this:
export BULB_IP=192.168.1.100
export BULB_PORT=4001
Bulb Configuration File
Create bulb_config.json:
{
"default_bulb": {
"ip": "192.168.1.45",
"port": 4000,
"timeout": 5.0
},
"discovery": {
"enabled": true,
"timeout": 10.0,
"port_range": {
"start": 4000,
"end": 4010
}
}
}
🔍 IntelliGlow vs Traditional Solutions
| Feature | Alexa/Google | IntelliGlow |
|---|---|---|
| Voice Control | ✅ Basic commands | ✅ Natural language + AI reasoning |
| AI Integration | ❌ Limited ecosystem | ✅ Works with any AI model (Claude, GPT, etc.) |
| Hardware Control | ❌ Cloud-dependent | ✅ Direct UDP networking |
| Customization | ❌ Vendor limitations | ✅ Full control over protocol |
| Context Understanding | ❌ Simple keywords | ✅ AI understands context and workflows |
| Privacy | ❌ Cloud processing | ✅ Local processing |
| Developer Freedom | ❌ Closed ecosystem | ✅ Open protocol, extensible |
Result: IntelliGlow = Convenience of Alexa + Intelligence of AI + Freedom of Open Source! 🎉
🧪 Testing
# Test real UDP communication with your bulb
python test_network_bulbs.py
This will:
- 🔌 Test direct connection to 192.168.1.45:4000
- 🔍 Scan network for other bulbs
- 🤖 Simulate AI/MCP commands
- 🎤 Test voice command processing
🐛 Troubleshooting
No Bulb Found
- Ensure your smart bulb is on the same network
- Check that the bulb is listening on port 4000
- Try network discovery:
python -c "import asyncio; from mcp_server_smartbulb.bulb_discovery import BulbDiscovery; asyncio.run(BulbDiscovery().discover_bulbs())"
Voice Not Working
- Install voice dependencies:
pip install -e .[voice] - Check microphone permissions
- Test with:
python -m mcp_server_smartbulb.voice_interface
Connection Timeout
- Check firewall settings
- Verify bulb IP address
- Increase timeout in
bulb_config.json
📁 Project Structure
IntelliGlow/
├── mcp_server_smartbulb/
│ ├── __init__.py # Package initialization
│ ├── network_server.py # Main UDP-enabled MCP server
│ ├── udp_client.py # UDP networking client
│ ├── bulb_discovery.py # Network discovery
│ └── voice_interface.py # Voice command processing
├── bulb_config.json # Network configuration
├── test_network_bulbs.py # UDP testing script
├── voice_enabled_server.py # Complete IntelliGlow system
├── README.md # This file
└── pyproject.toml # Project configuration
Clean, focused, and intelligent! 🧠💡
🎯 What Makes IntelliGlow Special
IntelliGlow isn't just another smart bulb controller - it's the bridge between AI intelligence and physical hardware.
🔥 Key Innovations:
- AI-Native Design: Built for AI reasoning, not just voice commands
- Open Protocol: Works with any AI model, not locked to one vendor
- Local Processing: Privacy-focused, no cloud dependency required
- Hybrid Interface: Voice + AI chat + MCP protocol
- Developer Freedom: Full customization and extensibility
🌟 Real-World Magic:
User: "I'm working late and need focus lighting"
IntelliGlow:
→ AI understands context
→ Sets cool white light (5000K)
→ Optimal brightness (85%)
→ Direct UDP communication
→ Responds with confirmation
This is the future of smart homes - lighting that truly understands and adapts to your needs! 🚀
Made with ❤️ for the next generation of intelligent home automation
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。