A community-driven registry for the Claude Code ecosystem. Not affiliated with Anthropic.
Are you the author? Sign in to claim
A MCP server for Home Assistant
The server uses the MCP protocol to share access to a local Home Assistant instance with an LLM application.
A powerful bridge between your Home Assistant instance and Language Learning Models (LLMs), enabling natural language control and monitoring of your smart home devices through the Model Context Protocol (MCP). This server provides a comprehensive API for managing your entire Home Assistant ecosystem, from device control to system administration.
The server includes a powerful Server-Sent Events (SSE) system that provides real-time updates from your Home Assistant instance. This allows you to:
const eventSource = new EventSource(
'http://localhost:3000/subscribe_events?token=YOUR_TOKEN&domain=light'
);
eventSource.onmessage = (event) => {
const data = JSON.parse(event.data);
console.log('Update received:', data);
};
See SSE_API.md for complete documentation of the SSE system.
Add-on Management
Package Management (HACS)
Automation Management
Intelligent Organization
Robust Architecture
# Clone the repository
git clone https://github.com/tevonsb/homeassistant-mcp.git
cd homeassistant-mcp
# Install dependencies
npm install
# Build the project
npm run build
The project includes Docker support for easy deployment and consistent environments across different platforms.
Clone the repository:
git clone https://github.com/tevonsb/homeassistant-mcp.git
cd homeassistant-mcp
Configure environment:
cp .env.example .env
Edit the .env file with your Home Assistant configuration:
# Home Assistant Configuration
HASS_HOST=http://homeassistant.local:8123
HASS_TOKEN=your_home_assistant_token
HASS_SOCKET_URL=ws://homeassistant.local:8123/api/websocket
# Server Configuration
PORT=3000
NODE_ENV=production
DEBUG=false
Build and run with Docker Compose:
# Build and start the containers
docker compose up -d
# View logs
docker compose logs -f
# Stop the service
docker compose down
Verify the installation:
The server should now be running at http://localhost:3000. You can check the health endpoint at http://localhost:3000/health.
Update the application:
# Pull the latest changes
git pull
# Rebuild and restart the containers
docker compose up -d --build
The Docker setup includes:
All environment variables can be configured in the .env file. The following variables are supported:
HASS_HOST: Your Home Assistant instance URLHASS_TOKEN: Long-lived access token for Home AssistantHASS_SOCKET_URL: WebSocket URL for Home AssistantPORT: Server port (default: 3000)NODE_ENV: Environment (production/development)DEBUG: Enable debug mode (true/false)# Home Assistant Configuration
HASS_HOST=http://homeassistant.local:8123 # Your Home Assistant instance URL
HASS_TOKEN=your_home_assistant_token # Long-lived access token
HASS_SOCKET_URL=ws://homeassistant.local:8123/api/websocket # WebSocket URL
# Server Configuration
PORT=3000 # Server port (default: 3000)
NODE_ENV=production # Environment (production/development)
DEBUG=false # Enable debug mode
# Test Configuration
TEST_HASS_HOST=http://localhost:8123 # Test instance URL
TEST_HASS_TOKEN=test_token # Test token
.env.example to .env.development.env.example to .env.production.env.example to .env.testTo use your new Home Assistant MCP server, you can add Claude Desktop as a client. Add the following to the configuration. Note this will run the MCP within claude and does not work with the Docker method.
{
"homeassistant": {
"command": "node",
"args": [<path/to/your/dist/folder>]
"env": {
NODE_ENV=development
HASS_HOST=http://homeassistant.local:8123
HASS_TOKEN=your_home_assistant_token
PORT=3000
HASS_SOCKET_URL=ws://homeassistant.local:8123/api/websocket
LOG_LEVEL=debug
}
}
}
{
"tool": "control",
"command": "turn_on", // or "turn_off", "toggle"
"entity_id": "light.living_room"
}
{
"tool": "control",
"command": "turn_on",
"entity_id": "light.living_room",
"brightness": 128,
"color_temp": 4000,
"rgb_color": [255, 0, 0]
}
{
"tool": "addon",
"action": "list"
}
{
"tool": "addon",
"action": "install",
"slug": "core_configurator",
"version": "5.6.0"
}
{
"tool": "addon",
"action": "start", // or "stop", "restart"
"slug": "core_configurator"
}
{
"tool": "package",
"action": "list",
"category": "integration" // or "plugin", "theme", "python_script", "appdaemon", "netdaemon"
}
{
"tool": "package",
"action": "install",
"category": "integration",
"repository": "hacs/integration",
"version": "1.32.0"
}
{
"tool": "automation_config",
"action": "create",
"config": {
"alias": "Motion Light",
"description": "Turn on light when motion detected",
"mode": "single",
"trigger": [
{
"platform": "state",
"entity_id": "binary_sensor.motion",
"to": "on"
}
],
"action": [
{
"service": "light.turn_on",
"target": {
"entity_id": "light.living_room"
}
}
]
}
}
{
"tool": "automation_config",
"action": "duplicate",
"automation_id": "automation.motion_light"
}
GET /api/state
POST /api/state
Manages the current state of the system.
Example Request:
POST /api/state
{
"context": "living_room",
"state": {
"lights": "on",
"temperature": 22
}
}
POST /api/context
Updates the current context with new information.
Example Request:
POST /api/context
{
"user": "john",
"location": "kitchen",
"time": "morning",
"activity": "cooking"
}
POST /api/action
Executes a specified action with given parameters.
Example Request:
POST /api/action
{
"action": "turn_on_lights",
"parameters": {
"room": "living_room",
"brightness": 80
}
}
POST /api/actions/batch
Executes multiple actions in sequence.
Example Request:
POST /api/actions/batch
{
"actions": [
{
"action": "turn_on_lights",
"parameters": {
"room": "living_room"
}
},
{
"action": "set_temperature",
"parameters": {
"temperature": 22
}
}
]
}
GET /api/actions
Returns a list of all available actions.
Example Response:
{
"actions": [
{
"name": "turn_on_lights",
"parameters": ["room", "brightness"],
"description": "Turns on lights in specified room"
},
{
"name": "set_temperature",
"parameters": ["temperature"],
"description": "Sets temperature in current context"
}
]
}
GET /api/context?type=current
Retrieves context information.
Example Response:
{
"current_context": {
"user": "john",
"location": "kitchen",
"time": "morning",
"activity": "cooking"
}
}
The server supports real-time updates via WebSocket connections.
// Client-side connection example
const ws = new WebSocket('ws://localhost:3000/ws');
ws.onmessage = (event) => {
const data = JSON.parse(event.data);
console.log('Received update:', data);
};
state_change: Emitted when system state changescontext_update: Emitted when context is updatedaction_executed: Emitted when an action is completederror: Emitted when an error occursExample Event Data:
{
"event": "state_change",
"data": {
"previous_state": {
"lights": "off"
},
"current_state": {
"lights": "on"
},
"timestamp": "2024-03-20T10:30:00Z"
}
}
All endpoints return standard HTTP status codes:
Error Response Format:
{
"error": {
"code": "INVALID_PARAMETERS",
"message": "Missing required parameter: room",
"details": {
"missing_fields": ["room"]
}
}
}
The API implements rate limiting to prevent abuse:
When rate limit is exceeded, the server returns:
{
"error": {
"code": "RATE_LIMIT_EXCEEDED",
"message": "Too many requests",
"reset_time": "2024-03-20T10:31:00Z"
}
}
# Get current state
curl -X GET \
http://localhost:3000/api/state \
-H 'Authorization: ApiKey your_api_key_here'
# Execute action
curl -X POST \
http://localhost:3000/api/action \
-H 'Authorization: ApiKey your_api_key_here' \
-H 'Content-Type: application/json' \
-d '{
"action": "turn_on_lights",
"parameters": {
"room": "living_room",
"brightness": 80
}
}'
// Execute action
async function executeAction() {
const response = await fetch('http://localhost:3000/api/action', {
method: 'POST',
headers: {
'Authorization': 'ApiKey your_api_key_here',
'Content-Type': 'application/json'
},
body: JSON.stringify({
action: 'turn_on_lights',
parameters: {
room: 'living_room',
brightness: 80
}
})
});
const data = await response.json();
console.log('Action result:', data);
}
# Development mode with hot reload
npm run dev
# Build project
npm run build
# Production mode
npm run start
# Run tests
npx jest --config=jest.config.cjs
# Run tests with coverage
npx jest --coverage
# Lint code
npm run lint
# Format code
npm run format
Node.js Version (toSorted is not a function)
nvm install 20.10.0
nvm use 20.10.0
Connection Issues
HASS_HOST accessibilityAdd-on Management Issues
HACS Integration Issues
Automation Issues
✅ Complete
🚧 In Progress
MIT License - See LICENSE file
Run Claude Code as an MCP server so any agent can delegate coding tasks to it
Browser automation using accessibility snapshots instead of screenshots
English-first Korean equity intelligence MCP — DART filings, foreign-holder 5%-rule flows, activist filings, KRX news. F
Unity MCP acts as a bridge between AI assistants and your Unity Editor. Give your LLM tools to manage assets, control sc
0
via CLI
