Overview
HA MCP is an open-source, high-performance solution engineered to solve foundational workflow challenges in the Python ecosystem. Built on local-first principles to ensure privacy, performance, and complete data ownership, it provides a unified architecture for modern developers and technical power users.
How It Works
HA MCP operates through structured, deterministic execution stages:
Home Assistant Instance Discovery: Connects to local Home Assistant core installations via authenticated REST and WebSocket APIs.
Tool Schema Registration: Automatically registers over 80 discrete tools spanning climate control, lights, security switches, and media players.
Agent Intent Orchestration: Translates natural language prompts from Claude or ChatGPT into precise Home Assistant service calls.
State Telemetry Verification: Reads entity state changes and sensor feedback in real time to verify successful automation execution.
Key Technical Features & Capabilities
High-Performance Architecture: Engineered on top of Python, ai, claude, minimizing runtime latency and resource overhead.
Local-First Reliability: Executes core workloads locally on host hardware with zero unexpected external telemetry or cloud vendor lock-in.
Extensible Modularity: Cleanly decoupled module boundaries enable rapid integration, custom scripting, and plugin ecosystem extensibility.
Standardized Developer Tooling: Incorporates automated dependency workflows, reproducible builds, and developer-friendly CLI hooks.
Architecture & Technology Stack
Primary Languages: Python, ai, claude
Build Systems & Runtimes: Python 3.10+ / uv / virtualenv (uv / pip / poetry)
Supported Platforms: macOS (Apple Silicon & Intel), Windows 10/11 x64, Linux
System Requirements
| Specification | Minimum Requirement | Recommended Specification |
|---|
| Operating System | macOS 12.0+ / Windows 10 x64 / Ubuntu 22.04 | macOS 14.0+ Sonoma / Windows 11 x64 |
| Processor Architecture | 4-Core x86_64 or Apple Silicon M1 | 8-Core Apple Silicon / Core i7+ |
| Memory (RAM) | 4 GB Unified Memory | 16 GB+ High-Speed RAM |
| Disk Space | 1 GB free SSD storage | 5 GB+ free SSD storage |
Installation & Setup Guide
macOS Installation
1Download verified release installer from CookApps or upstream releases.
2Drag application package into /Applications.
Windows Installation
1Download official setup binary (.exe / .msi).
2Run installer and launch from Start Menu.
Building from Source
git clone https://github.com/homeassistant-ai/ha-mcp.git
cd ha-mcp
pip install -e .
How to Use & Workflow
1Application Launch: Launch HA MCP to initialize runtime state and local storage caches.
2Configuration & Workspace: Set up local configuration, environment flags, or CLI paths.
3Execution Workflow: Execute automated pipelines or interact through the visual interface.
4Data Verification: Review generated outputs stored deterministically in designated directories.
Troubleshooting & Common Issues
Issue: Port or Resource Collision on Startup.
*Cause*: A zombie process or background daemon is holding active port bindings.
*Resolution*: Terminate holding processes before relaunching the application.
Issue: Missing native toolchain dependencies during build.
*Cause*: Host system lacks C++ or platform compilation tools.
*Resolution*: Install Xcode Command Line Tools (macOS) or Visual Studio Build Tools (Windows).