---
id: ha-mcp
version: "8.2.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# ha-mcp — Home Assistant MCP Server - Complete control of Home Assistant through MCP

License: permissive · Maintenance: active · Downloads: 427.6K/mo

## What it is and what it does

ha-mcp is a Model Context Protocol server that bridges AI assistants and Home Assistant. It exposes Home Assistant's device control, state queries, service calls, and automation management through an MCP interface, allowing AI models to interact with smart homes using natural language. The package can run as a Home Assistant custom component (in-process, recommended), a standalone add-on, Docker container, or via PyPI/uvx as an HTTP server. It depends on fastmcp for MCP protocol handling, httpx for HTTP requests, pydantic for data validation, python-dotenv for environment configuration, cryptography and truststore for secure connections, and packaging utilities.

The server supports multiple deployment modes and authentication strategies—webhook-based remote access through Nabu Casa or reverse proxies, local-only operation, optional Home Assistant account sign-in, and OIDC integration for external identity providers. A Home Assistant sidebar panel allows administrators to manage tools, feature flags, and settings. The package is actively maintained (last commit 2026-08-14) and marked Production/Stable.

Use it for:

- Control smart home devices (lights, locks, thermostats) via natural language through Claude, ChatGPT, or other MCP-compatible AI clients.
- Query device states and automations from an AI assistant without manual Home Assistant dashboard navigation.
- Execute Home Assistant services and automations through conversational AI on local network or remote access.
- Integrate Home Assistant into AI workflows for home automation decision-making and multi-step tasks.
- Deploy ha-mcp in Home Assistant OS, Supervised, Container, or Core installations with a single install method.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides a Model Context Protocol (MCP) server that connects AI assistants to Home Assistant, enabling natural-language control of smart home devices, state queries, service execution, and automation management.

Yes. Active maintenance, low install friction, permissive MIT license, no known vulnerabilities, and broad Python 3.13–3.14 support make it a solid choice. The custom component integration path is particularly smooth for Home Assistant users. Install if you want AI assistants to control your smart home; the recommended in-process setup requires no separate token management.

## Install

pip install ha-mcp
uv add ha-mcp
poetry add ha-mcp

## Installing ha-mcp

Before you install: Low friction: pure Python wheel with no compiled dependencies. Active maintenance—last release 4 days ago, repository at 4381 stars. Requires Python 3.13 or 3.14 only; supports current versions.

License in practice: MIT license (permissive): you may use, modify, and distribute freely with minimal restrictions. No copyleft obligations.

Quickstart:

pip install ha-mcp

ha-mcp --help

# Or via uvx:
uvx ha-mcp@latest

Requires Python 3.13 or 3.14 (requires_python: <3.15,>=3.13). Home Assistant instance and access token needed for actual operation.

Verify before relying:

- Performance characteristics under concurrent AI client connections.
- Full feature parity across all Home Assistant installation types (OS, Supervised, Container, Core) in practice.
- Stability of the in-process custom component on long-running Home Assistant instances.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 427.6K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

home assistant mcp server, ai smart home control, natural language home automation, home assistant ai integration, mcp home assistant, smart home voice assistant bridge, home assistant api for ai, home-automation, mcp-protocol, ai-integration

[View on SkillFed](https://skillfed.io/packages/ha-mcp) · [View on PyPI](https://pypi.org/project/ha-mcp/)