Skip to main content
Glama
GETTING_STARTED.md9.95 kB
# 🎯 Getting Started with Home Assistant MCP Welcome to Home Assistant MCP! This guide will help you get up and running with AI-powered smart home control in just a few minutes. ## 📋 Table of Contents - [What is Home Assistant MCP?](#what-is-home-assistant-mcp) - [Prerequisites](#prerequisites) - [Quick Installation](#quick-installation) - [First Steps](#first-steps) - [Basic Usage](#basic-usage) - [Next Steps](#next-steps) --- ## What is Home Assistant MCP? Home Assistant MCP (Model Context Protocol) server enables AI assistants like Claude, GPT-4, and Cursor to control your smart home through natural language. Simply tell your AI assistant what you want, and it will: - ✅ Turn lights on/off and adjust brightness - ✅ Control thermostats and climate devices - ✅ Manage media players and entertainment - ✅ Trigger automations and scenes - ✅ Monitor your home's status - ✅ Analyze energy usage and optimize efficiency **Example conversation:** ``` You: "Turn on the living room lights and set them to 50% brightness" AI: *Executes command through Home Assistant MCP* AI: "Done! The living room lights are now on at 50% brightness." ``` --- ## Prerequisites Before you begin, make sure you have: ### 1. 🏠 Home Assistant Instance You need a running Home Assistant instance: - **Home Assistant OS**: Install from [home-assistant.io](https://www.home-assistant.io/installation/) - **Home Assistant Container**: Docker-based installation - **Home Assistant Core**: Python-based installation - **Home Assistant Supervised**: For advanced users **Minimum Version**: Home Assistant 2023.1 or newer ### 2. 🔑 Long-Lived Access Token Create a token for secure API access: 1. Open Home Assistant 2. Click your profile (bottom left) 3. Scroll to "Long-Lived Access Tokens" 4. Click "Create Token" 5. Give it a name (e.g., "MCP Server") 6. Copy and save the token securely > **⚠️ Important**: Keep your token secure! It provides full access to your Home Assistant instance. ### 3. 🚀 Runtime Environment Choose one of the following: - **Option A**: [Bun](https://bun.sh) v1.0.26+ (Recommended - fastest) - **Option B**: [Node.js](https://nodejs.org/) v18.0.0+ - **Option C**: [Docker](https://www.docker.com/) **Why Bun?** It's 4x faster than Node.js and has built-in TypeScript support. ### 4. 🤖 AI Assistant (MCP Client) Install one of these AI assistants: - **Claude Desktop**: [Download here](https://claude.ai/download) - **Cursor**: [Download here](https://cursor.sh/) - **VS Code + Claude Extension**: Available in VS Code marketplace --- ## Quick Installation ### Method 1: Smithery (Easiest - Recommended for Beginners) [Smithery](https://smithery.ai) provides one-click installation: ```bash # For Claude Desktop npx @smithery/cli install @jango-blockchained/homeassistant-mcp --client claude # For Cursor npx @smithery/cli install @jango-blockchained/homeassistant-mcp --client cursor # For VS Code npx @smithery/cli install @jango-blockchained/homeassistant-mcp --client vscode ``` **What happens:** 1. You'll be prompted for your Home Assistant URL (e.g., `http://192.168.1.100:8123`) 2. Paste your long-lived access token 3. Configure optional settings 4. Installation completes automatically **Restart your AI assistant** and you're ready to go! ### Method 2: NPX (Quick Start) Run directly without installation: ```bash npx @jango-blockchained/homeassistant-mcp@latest ``` **First-time setup:** 1. Create a `.env` file in your project directory 2. Add your credentials: ```env HOME_ASSISTANT_URL=http://192.168.1.100:8123 HOME_ASSISTANT_TOKEN=your_long_lived_token_here ``` ### Method 3: Docker (Containerized) For isolated, containerized deployment: ```bash # Pull the image docker pull ghcr.io/jango-blockchained/advanced-homeassistant-mcp:latest # Run the container docker run -d \ --name homeassistant-mcp \ -e HOME_ASSISTANT_URL=http://192.168.1.100:8123 \ -e HOME_ASSISTANT_TOKEN=your_token_here \ -p 4000:4000 \ ghcr.io/jango-blockchained/advanced-homeassistant-mcp:latest ``` ### Method 4: From Source (For Developers) Clone and build from source: ```bash # Clone the repository git clone https://github.com/jango-blockchained/advanced-homeassistant-mcp.git cd advanced-homeassistant-mcp # Install dependencies bun install # Configure environment cp .env.example .env # Edit .env with your settings # Build the project bun run build:all # Start the server bun run start:stdio ``` --- ## First Steps ### 1. Verify Installation After installation, verify the MCP server is available: **For Claude Desktop:** 1. Open Claude Desktop 2. Look for the 🔌 icon (bottom right) 3. You should see "homeassistant-mcp" listed **For Cursor:** 1. Open Cursor 2. Press `Cmd/Ctrl + Shift + P` 3. Type "MCP" and check available servers **For VS Code:** 1. Open VS Code 2. Check the MCP extension status bar ### 2. Test Basic Commands Try these simple commands with your AI assistant: **List your devices:** ``` "Show me all my smart home devices" ``` **Control a light:** ``` "Turn on the bedroom light" "Set the living room lights to 30%" ``` **Check status:** ``` "What's the temperature in the living room?" "Which lights are currently on?" ``` ### 3. Explore Capabilities Ask your AI assistant: ``` "What can you do with my smart home?" "Show me examples of home automation commands" "Help me understand the available features" ``` --- ## Basic Usage ### Controlling Lights ``` 💬 "Turn off all bedroom lights" 💬 "Set the kitchen lights to warm white at 70% brightness" 💬 "Make the living room lights blue" 💬 "Dim all lights to 20%" ``` ### Managing Climate ``` 💬 "Set the thermostat to 72°F" 💬 "Turn on heating in the bedroom" 💬 "What's the current temperature in the living room?" 💬 "Set all thermostats to eco mode" ``` ### Media Control ``` 💬 "Play music on the living room speaker" 💬 "Pause all media players" 💬 "Set the TV volume to 50%" 💬 "Switch to HDMI 2 on the bedroom TV" ``` ### Automation & Scenes ``` 💬 "Activate the movie night scene" 💬 "Trigger my morning routine" 💬 "Show me all my automations" 💬 "Create a bedtime scene" ``` ### Smart Features ``` 💬 "Check my home's health status" 💬 "Find devices that are offline" 💬 "Show me my energy usage" 💬 "Are there any issues I should know about?" ``` --- ## Understanding Responses ### Success Response Example ``` ✅ Successfully turned on light.living_room Brightness: 128 (50%) Color temperature: 370 (warm white) State: on ``` ### Error Response Example ``` ❌ Error: Entity light.nonexistent not found Suggestion: Run "list all lights" to see available lights ``` ### Information Response Example ``` 📊 Living Room Status: • Temperature: 72°F • Humidity: 45% • Lights: 2 on, 3 off • Media: Currently playing ``` --- ## Common Patterns ### Morning Routine ``` 💬 "Good morning! Can you: - Turn on the bedroom lights to 50% - Set the thermostat to 72°F - Start my morning playlist - Open the living room blinds" ``` ### Leaving Home ``` 💬 "I'm leaving home. Please: - Turn off all lights - Set thermostats to eco mode - Lock all doors - Activate the security alarm" ``` ### Energy Saving ``` 💬 "Help me save energy: - Find lights that are on but not needed - Check for heating conflicts with open windows - Show devices consuming the most power - Suggest optimizations" ``` --- ## Troubleshooting Quick Tips ### Issue: AI assistant can't see the MCP server **Solution:** 1. Restart your AI assistant (Claude, Cursor, etc.) 2. Check that the server is running 3. Verify the configuration file is correct ### Issue: Commands fail with "unauthorized" error **Solution:** 1. Verify your Home Assistant token is valid 2. Check that the token hasn't expired 3. Ensure the token has proper permissions ### Issue: Devices not found **Solution:** 1. Verify devices are available in Home Assistant 2. Check device entity IDs: `"List all my lights"` 3. Ensure Home Assistant is accessible from the MCP server ### Issue: Slow responses **Solution:** 1. Check your network connection to Home Assistant 2. Verify Home Assistant is responsive 3. Consider running the MCP server closer to Home Assistant For more detailed troubleshooting, see the [Troubleshooting Guide](TROUBLESHOOTING.md). --- ## Next Steps Now that you're up and running, explore these resources: ### 📚 For Users 1. **[Quick Start Scenarios](QUICK_START_SCENARIOS.md)** - Real-world usage examples 2. **[Tools Reference](TOOLS_REFERENCE.md)** - Complete list of available commands 3. **[Smart Features](SMART_FEATURES.md)** - Advanced maintenance and automation 4. **[Configuration Guide](CONFIGURATION.md)** - Fine-tune your setup ### 🔧 For Developers 1. **[Architecture Guide](ARCHITECTURE.md)** - System design and internals 2. **[API Reference](API_REFERENCE.md)** - Programmatic access 3. **[Development Guide](DEVELOPMENT.md)** - Contributing and extending 4. **[Testing Guide](TESTING.md)** - Testing your changes ### 🎓 Advanced Topics 1. **[Docker Deployment](DOCKER_GUIDE.md)** - Production deployment 2. **[Security Best Practices](SECURITY.md)** - Securing your setup 3. **[Performance Tuning](PERFORMANCE.md)** - Optimization tips 4. **[FAQ](FAQ.md)** - Frequently asked questions --- ## Need Help? - 💬 [GitHub Discussions](https://github.com/jango-blockchained/advanced-homeassistant-mcp/discussions) - Ask questions - 🐛 [Issue Tracker](https://github.com/jango-blockchained/advanced-homeassistant-mcp/issues) - Report bugs - 📖 [Documentation](index.md) - Browse all docs - 🏠 [Home Assistant Community](https://community.home-assistant.io/) - General HA help --- **Ready to transform your smart home?** Start experimenting with commands and see what your AI assistant can do! 🚀

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/jango-blockchained/advanced-homeassistant-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server