homeassistant-mcp
Enables AI assistants to control Home Assistant devices and services through the Model Context Protocol.
README
Home Assistant MCP Server
A Model Context Protocol (MCP) server that lets AI assistants control Home Assistant. Built with Python and FastMCP.
Works with Claude, GPT-4, Cursor, Kiro, and any MCP-compatible client.
What it does
- 40 tools covering lights, climate, covers, locks, media players, vacuums, fans, cameras, alarms, and more
- BM25 tool search — LLMs discover tools on demand instead of receiving all 40 schemas upfront
- MCP Resources for read-only entity, area, device, and service data
- MCP Prompts for guided workflows (automation creation, troubleshooting, energy optimization)
- Tool annotations —
readOnlyHintlets clients skip confirmation prompts for safe operations - Smart error log parsing — deduplicates and summarises HA error logs instead of dumping raw text
- Context-aware responses — compact list responses, domain-filtered queries, progress reporting
- Async throughout with TTL-based caching to reduce API load
Installation
uvx (recommended)
No install needed. Just configure your MCP client:
{
"mcpServers": {
"homeassistant": {
"command": "uvx",
"args": ["homeassistant-mcp"],
"env": {
"HASS_HOST": "http://homeassistant.local:8123",
"HASS_TOKEN": "your_long_lived_access_token_here"
}
}
}
}
pip
pip install homeassistant-mcp
homeassistant-mcp
From source
git clone https://github.com/robbrad/homeassistant-mcp.git
cd homeassistant-mcp
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -e ".[dev]"
homeassistant-mcp
Configuration
The server needs two environment variables:
| Variable | Description |
|---|---|
HASS_HOST |
Home Assistant URL, e.g. http://homeassistant.local:8123 |
HASS_TOKEN |
Long-lived access token (how to create one) |
Optional:
| Variable | Default | Description |
|---|---|---|
CACHE_TTL_STATES |
30 |
Cache TTL for bulk state queries (seconds) |
CACHE_TTL_ENTITY |
10 |
Cache TTL for individual entity queries (seconds) |
LOG_LEVEL |
INFO |
Logging level |
These can be set via environment variables (as shown in the MCP client configs above) or in a .env file. See .env.example.
MCP Client Setup
Claude Desktop
Add to your config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"homeassistant": {
"command": "uvx",
"args": ["homeassistant-mcp"],
"env": {
"HASS_HOST": "http://homeassistant.local:8123",
"HASS_TOKEN": "your_token_here"
}
}
}
}
Cursor / Kiro / Other MCP Clients
Same config format. Place it in your client's MCP configuration file (e.g. .cursor/mcp.json, .kiro/settings/mcp.json).
The server uses stdio transport, which is the standard for MCP.
How Tool Discovery Works
With 40 tools, the server uses BM25 search to keep the LLM's context lean. Instead of sending all 40 tool schemas upfront, the LLM sees 8 tools:
| Tool | Purpose |
|---|---|
discover_tools |
Returns the full catalog of all 40 tools by category |
states_control |
Entity state management (list, get, set, delete) |
list_devices |
Device discovery with domain/area/floor filtering |
call_service |
Call any HA service directly |
template_render |
Render Jinja2 templates in HA context |
error_log_get |
Parsed and deduplicated error log summary |
search_tools |
BM25 search to find tools by description |
call_tool |
Execute a discovered tool by name |
The LLM calls discover_tools() to see what's available, then search_tools(query="lights brightness") to get the full schema, then calls the tool directly.
All Tools
Device Control (16 tools)
| Tool | Domain | Capabilities |
|---|---|---|
lights_control |
light | Brightness, color temp, RGB |
climate_control |
climate | HVAC modes, temperature, fan |
switch_control |
switch | On/off, bulk operations |
cover_control |
cover | Position, tilt, open/close |
lock_control |
lock | Lock/unlock with codes |
media_player_control |
media_player | Playback, volume, source |
camera_control |
camera | Snapshots, streams, motion detection |
vacuum_control |
vacuum | Start, dock, fan speed |
fan_control |
fan | Speed, oscillation, direction |
alarm_control |
alarm_control_panel | Arm/disarm modes |
weather_control |
weather | Conditions, daily/hourly forecasts |
water_heater_control |
water_heater | Temperature, modes |
humidifier_control |
humidifier | Humidity levels |
siren_control |
siren | Activation control |
valve_control |
valve | Open/close |
lawn_mower_control |
lawn_mower | Start, stop, dock |
Automation & Scenes (3 tools)
| Tool | Capabilities |
|---|---|
automation_control |
List, trigger, enable, disable, reload |
scene_control |
List, activate |
script_control |
List, execute with variables, reload |
Input Helpers (5 tools)
| Tool | Capabilities |
|---|---|
input_boolean_control |
Toggle on/off |
input_number_control |
Set value, increment, decrement |
input_select_control |
Select from options |
input_text_control |
Set text value |
input_datetime_control |
Set date/time |
API & State (4 tools)
| Tool | Capabilities |
|---|---|
api_info |
API status, HA config, loaded components |
events_control |
List event types, fire custom events |
services_control |
List services by domain, call services |
states_control |
List/get/set/delete entity states with filtering |
History (3 tools)
| Tool | Capabilities |
|---|---|
history_query |
State changes by entity + hours (not ISO timestamps) |
logbook_query |
Human-readable logbook entries |
error_log_get |
Parsed, deduplicated error summary with component counts |
Specialized (5 tools)
| Tool | Capabilities |
|---|---|
calendar_access |
List calendars, get events by date range |
camera_proxy_get |
Camera images with optional resize |
config_check |
Validate HA configuration |
intent_handle |
Process natural language intents |
template_render |
Render Jinja2 templates |
General (4 tools)
| Tool | Capabilities |
|---|---|
list_devices |
Filter by any domain, area, or floor |
call_service |
Call any HA service with custom data |
send_notification |
Send alerts via HA notification services |
discover_tools |
Full tool catalog for LLM discovery |
MCP Resources
| URI Pattern | Description |
|---|---|
hass://entity/{entity_id} |
Entity state and attributes |
hass://area/{area_id} |
Area entities (compact summaries) |
hass://device/{device_id} |
Device entities (compact summaries) |
hass://services |
All services organized by domain |
hass://entity/{entity_id}/history |
Entity history with pagination |
Development
git clone https://github.com/robbrad/homeassistant-mcp.git
cd homeassistant-mcp
python -m venv venv
source venv/bin/activate
pip install -e ".[dev]"
# Run tests
pytest
# Format and lint
black src/ tests/
ruff check src/ tests/
# Type check
mypy src/
Publishing
Commits to main auto-publish to PyPI via GitHub Actions + python-semantic-release:
fix:commits → patch bump (3.1.0 → 3.1.1)feat:commits → minor bump (3.1.0 → 3.2.0)BREAKING CHANGE:→ major bump (3.1.0 → 4.0.0)
Architecture
AI Assistant <--> MCP Server (FastMCP/stdio) <--> Home Assistant REST API
|
+-----+-----+
| |
Tools Cache
Layer Layer
- FastMCP Server handles MCP protocol over stdio with BM25 tool search
- Home Assistant Client is an async httpx client with auth and connection pooling
- Cache Layer provides TTL-based caching for states and entities
- Tools Layer — 40 tools with annotations, tags, timeouts, and context logging
- Resources — 5 read-only data endpoints with compact response envelopes
- Prompts — 13 guided workflows for automation, troubleshooting, and optimization
License
MIT
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。