Build

Custom Software DevelopmentSystems built to your process

Web Application DevelopmentReact, Next.js, Rails, Django

Mobile App DevelopmentiOS, Android, React Native

SaaS Product DevelopmentMulti-tenant from day one

MVP DevelopmentFirst release in 6–10 weeks

Enterprise SoftwareSSO, audit trails, scale

Connect & operate

API Development & IntegrationREST, GraphQL, webhooks

Cloud & DevOps EngineeringAWS, GCP, Azure, Terraform

QA & Software TestingAutomated suites, load, E2E

Legacy ModernizationStrangler-fig migrations

Cybersecurity EngineeringHardening and remediation

Data EngineeringPipelines, warehouses, reporting

Design & teams

UI/UX DesignInterfaces people finish tasks in

Prototyping & DiscoveryScope before you spend

Staff AugmentationSenior engineers on your team

Dedicated Dev TeamsA full pod, one contract

White-Label DevelopmentEngineering under your brand

Technology ArchitectureDecide once, correctly

Project ManagementDelivery you can audit

Start hereScope a build in one callBring a spec, a wireframe, or a paragraph. You leave with an architecture, a timeline, and a number.Book a scoping call

All software development services →How we price work →Talk to an engineer →

AI software

AI Application DevelopmentFull-stack AI products

AI Agent DevelopmentMulti-step autonomous agents

AI SaaS DevelopmentMulti-tenant AI products

Workflow AutomationReplace manual ops

Custom LLM AppsBranded internal assistants

LLM & data systems

Private LLM DevelopmentSelf-hosted Llama, Mistral, Qwen

RAG DevelopmentCited answers over your docs

MCP & WebMCPMake your systems agent-callable

Vector Database Setuppgvector, Pinecone, Qdrant

Document IntelligenceOCR and extraction pipelines

Vibe coding

AI-Assisted BuildsFast, with senior review

Rapid PrototypingClickable in days

Concept to CodeIdea to working software

Code Review & CleanupMake AI output shippable

Fractional AI TeamOngoing build capacity

Ready to ship?Put AI where the work isAgents, RAG, and private LLMs wired into the systems your team already uses — not a chatbot bolted to a homepage.Discuss an AI project

AI development services →AI readiness checklist →AI vs. traditional development →

Commercial

Financial Services

eCommerce & Retail

Marketing Agencies

Real Estate

Automotive

Regulated

Healthcare & HIPAA

Legal

Government

Education

Banking

By company size

Startups & MVPs

Small Business

Enterprise

Agriculture

Domain firstWe learn your workflow before we model itRegulated, operational, or high-volume — the constraints belong in the schema, not in a training doc.Talk about your domain

Selected workSoftware we own and operateMost agencies show screenshots. We run the products — the same stack, uptime, and on-call we hand to clients.View the portfolio

The DEV.co product network

AI ERP with agents inside the systems you run

AI phone and voice agents for real calls

LLM.coPrivate, self-hosted LLM deployments

Legal AI infrastructure for firms

AI RFP discovery and response drafting

Automatic.coBusiness process automation

CyberAttack.aiThreat detection and defense

Secure AI virtual data rooms

Compare

Custom Software vs. SaaS

In-House vs. Outsourced

MVP vs. Full Product

AI vs. Traditional Development

Tools & reading

Software Cost Calculator

Project Scope Template

Engineering Blog

Open Source

Plan smarterEstimate before you commitCost ranges, scope templates, and the questions we ask in discovery — free, no form.Open the cost calculator

DEV.co

About

Careers

Locations

Contact

Real conversationsTalk with a technical leadNo SDR, no discovery gauntlet. The person on the call is the one who scopes the build.Book a call

MCP Servers · homeassistant-ai

ha-mcp

ha-mcp is a Python-based Model Context Protocol server that bridges AI assistants (Claude, ChatGPT) with Home Assistant smart home systems. It exposes 85+ tools for controlling devices, querying states, and managing automations through natural language.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/homeassistant-ai/ha-mcp

3.9k

GitHub stars

152

Forks

Python

Primary language

MIT

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	homeassistant-ai/ha-mcp
Owner	homeassistant-ai
Primary language	Python
License	MIT — OSI-approved
Stars	3.9k
Forks	152
Open issues	10
Latest release	v7.10.0 (2026-07-06)
Last updated	2026-07-07
Source	https://github.com/homeassistant-ai/ha-mcp

What ha-mcp is

Built on FastMCP, ha-mcp translates MCP tool calls into Home Assistant service calls and state queries. Deployable as a HACS custom component (in-process), Home Assistant add-on, Docker container, or local Python package; supports HTTP streaming transport and optional webhook-based remote access.

Quickstart

Get the ha-mcp source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/homeassistant-ai/ha-mcp.git2cd ha-mcp3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI-driven smart home control via Claude Desktop or ChatGPT

Teams using Claude or ChatGPT can issue natural-language commands to control lights, thermostats, locks, and automations without leaving their chat interface. The component handles authentication via Home Assistant's existing user system.

Home Assistant installations needing AI agent integration

Home Assistant OS, Supervised, Container, and Core deployments can enable MCP without managing separate tokens or infrastructure. The custom component runs in-process, eliminating external dependencies and simplifying credential rotation.

Multi-client AI orchestration for home automation

Organizations running multiple AI assistants (Claude, ChatGPT, custom LLM-based tools) can point them all at a single ha-mcp endpoint, ensuring consistent device state visibility and preventing conflicting automations.

Implementation considerations

Choose one install method per client to avoid connection hangs; the custom component (HACS) is recommended for simplicity and requires no token management.
Enable webhook authentication (`ha_auth`) for multi-user setups to avoid sharing a secret URL; local-only mode is available if remote access is not needed.
Monitor the 10 open issues and breaking changes (e.g., v7.3.0 moved `ha_config_set_yaml` to beta); test upgrades in non-critical environments first.
Review the 85+ exposed tools in the settings panel to disable unused ones (especially file/YAML editing) if your organization has strict change-control policies.
Stdio transport is not production-ready; use HTTP (Docker/PyPI) or the custom component instead.

When to avoid it — and what to weigh

Your Home Assistant lacks network connectivity — ha-mcp requires a running Home Assistant instance. Offline-only smart home setups cannot use this server.
You need guaranteed security certifications or compliance audits — This is an unofficial, community-maintained project with no published security audit or SLA. Production IoT environments with regulatory requirements should review or fork the code independently.
You cannot expose any endpoint (local or remote) — Even local stdio mode has documented connection issues. All deployment methods require either a local HTTP port or a webhook URL, making fully isolated setups impractical.
Your AI client does not support MCP — Only Claude Desktop, Claude.ai, ChatGPT with MCP plugins, and MCP-compatible clients can use ha-mcp. Older chatbots or custom systems require additional integration work.

License & commercial use

MIT License. Permits use, modification, and distribution for commercial and private purposes with attribution and no warranty.

MIT is a permissive OSI license that explicitly allows commercial use. However, this is an unofficial, unmaintained-by-Home-Assistant project; commercial deployments should review the code, test thoroughly, and consider forking or contracting support, as the project has no SLA or guaranteed support channel.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Low
DEV.co fit	Strong
Assessment confidence	High

Security considerations

The custom component uses Home Assistant's native user authentication when `ha_auth` is enabled, avoiding token exposure. However, the secret URL mode (default) requires protecting the webhook URL; ensure Home Assistant's HTTPS and Nabu Casa or reverse proxy are configured correctly. The project is not professionally audited; teams handling sensitive devices (locks, cameras) should review the code. Avoid enabling file/YAML editing tools unless necessary.

Alternatives to consider

Home Assistant Cloud API + custom LLM integration

Home Assistant Cloud exposes a REST API; custom integration code can call it without MCP. Suited for teams with in-house LLM expertise and no need for Claude/ChatGPT integration.

Home Assistant Automations + Node-RED

Traditional automation rules and visual flows require no LLM or MCP. Ideal for deterministic, rule-based use cases; does not enable natural-language control.

OpenAI GPT home integration plugins (ChatGPT ecosystem)

ChatGPT plugins for smart home (if available for your vendor). Tightly coupled to ChatGPT but may not require running an external MCP server; check availability for your devices.

Software development agency

Build on ha-mcp with DEV.co software developers

Start with the HACS custom component (no token needed) or explore Docker/PyPI deployments. Use the setup wizard to configure your AI client in minutes.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

mcp mcp mcp-use

Related on DEV.co

Explore the category and the services that help you build with it.

MCP Servers AI Development Services AI-powered smart home integrations Home automation and IoT backend services Custom smart home orchestration Talk to DEV.co

ha-mcp FAQ

Do I need a token to use the custom component?

No. The custom component runs in-process inside Home Assistant and uses the Home Assistant user system for auth (if `ha_auth` is enabled) or a secret webhook URL. Token management is optional only for Docker/PyPI deployments.

Can I use ha-mcp with ChatGPT, or only Claude?

ChatGPT supports MCP via plugins. The setup wizard provides client-specific config. However, Claude Desktop and Claude.ai have been more thoroughly tested by the community.

What happens if I install ha-mcp two different ways (custom component + add-on)?

The README warns this causes connection hangs. Choose one install method per AI client and point the client only at that single URL.

Is the server actively maintained?

Yes, as of July 2026. E2E tests pass, releases are regular, and issues are addressed. However, there is no published SLA or guaranteed support—it is community-driven.

Work with a software development agency

Evaluating ha-mcp is the easy part; running it is not. DEV.co offers software development services and AI software development services for mcp servers teams that need a partner past the proof of concept.

Ready to add AI control to your smart home?

Start with the HACS custom component (no token needed) or explore Docker/PyPI deployments. Use the setup wizard to configure your AI client in minutes.

Talk to DEV.co Schedule a Consultation