About smarthome-port
smarthome-port provides CLI and MCP access to a Home Assistant instance through its REST API. It covers entity states and services as well as history, configuration, templates, and events.
Key features
- 01List entities and read entity states
- 02Call and list domain services
- 03Retrieve entity history and core configuration
- 04Render Jinja templates
- 05Fire events on the event bus
- 06CLI and stdio MCP server
Use cases
- 01Check the state of a Home Assistant entity
- 02Call a service such as turning on a light
- 03Review recent entity history
- 04Render a template or fire an event
MCP support
Exposes Home Assistant REST API operations as tools through a stdio MCP server. Configuration requires the instance URL and a long-lived access token.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @reneturcios/smarthome-portLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
8Saved tool list · updated 1w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| domain | string | No | Optional domain (the part of an entity_id before the dot, e.g. 'light') to filter listed entities to. |
Details
From the project READMEsmarthome-port
smarthome-port talks to a Home Assistant instance through its
REST API. Everything in Home Assistant is an entity —
light.kitchen, sensor.temp, switch.porch — and the API
reflects that shape: you read entity states, call services on a
domain (light.turn_on, switch.toggle), look back at recent
history, read the core configuration, render Jinja templates on
the server, and fire events onto the event bus. smarthome-port
exposes those six operations, plus listings of entities and
services, through one binary that serves a command-line interface
and a stdio MCP server.
Two settings are required, and there is no built-in default for either:
HOME_ASSISTANT_URL— the base URL of the instance, for examplehttp://homeassistant.local:8123. A trailing slash is stripped.HOME_ASSISTANT_TOKEN— a long-lived access token (create one in Home Assistant under your profile). It is sent as anAuthorization: Bearerheader and never appears in output.
When either variable is missing, every call fails with an error that names the missing variable, before any request is made.
Only the REST API is covered. The websocket API is not implemented.
Install
npx @reneturcios/smarthome-port entities
Or run the container image:
docker run -i -e HOME_ASSISTANT_URL -e HOME_ASSISTANT_TOKEN ghcr.io/cloudcorprecords/smarthome-port
The container defaults to the MCP server (see below). Pass CLI arguments to use it as a CLI instead.
CLI
export HOME_ASSISTANT_URL=http://homeassistant.local:8123
export HOME_ASSISTANT_TOKEN=<your long-lived access token>
smarthome-port entities
smarthome-port entities --domain light
smarthome-port state sensor.temp
smarthome-port call light turn_on --entity light.kitchen --data '{"brightness_pct":50}'
smarthome-port services
smarthome-port services --domain light
smarthome-port history sensor.temp --hours 6
smarthome-port config
smarthome-port render "{{ states('sensor.temp') }}"
smarthome-port fire my_event --data '{"device":"kitchen"}'
smarthome-port --version
Each command prints the upstream JSON payload (template rendering
prints {"rendered": "..."}). Exit codes: 0 on success, 1
when a call fails (including missing configuration or an API
error), 2 for usage errors.
MCP server
One bin serves both faces. The mcp subcommand starts a stdio MCP
server:
npx -y @reneturcios/smarthome-port mcp
MCP client configuration:
{
"mcpServers": {
"smarthome-port": {
"command": "npx",
"args": ["-y", "@reneturcios/smarthome-port", "mcp"],
"env": {
"HOME_ASSISTANT_URL": "http://homeassistant.local:8123",
"HOME_ASSISTANT_TOKEN": "..."
}
}
}
}
Tools
| Tool | Input | What it returns |
|---|---|---|
list_entities | domain (optional) | Entity states, filtered client-side to one domain when given. |
get_entity_state | entity_id (required) | The full state of one entity. |
call_service | domain, service (required), entity_id, service_data (optional) | The service call result; service_data is merged with entity_id. |
list_services | domain (optional) | Services by domain, or only one domain's entry when given. |
get_history | entity_id (required), hours (optional, default 24) | Recent history for the entity over the window ending now. |
get_config | (none) | The core configuration. |
render_template | template (required) | {"rendered": "..."} — the upstream reply is plain text. |
fire_event | event_type (required), event_data (optional) | The upstream reply; event_data defaults to {}. |
All calls go to {HOME_ASSISTANT_URL}/api. An entity_id must
contain a dot (domain.object_id), and domain, service, and
event_type must be non-empty — both are checked before any
request. History timestamps are computed client-side from the
current time. HTTP failures surface the status plus the upstream
message when Home Assistant provides one.
Development
npm install
npm run build # tsc -> dist/
npm test # node test runner over test/*.test.ts, no network (fake fetch fixtures)
npm run dev # run the CLI from source via tsx
License
MIT — see LICENSE.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.