cl0ud6uru

The AI that remembers, learns, and actually does things in your Home Assistant environment.

18
1
100% credibility
Found Apr 28, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

HestiaClaw is a self-hosted web-based AI assistant that integrates deeply with Home Assistant for natural language control of smart home devices, persistent memory via a 3D knowledge graph, voice input/output, and custom skills.

How It Works

1
🔍 Discover Hestia

You hear about Hestia, a friendly AI helper that lives on your own computer and chats with you about your smart home.

2
📥 Bring her home

Download the ready-to-go package and start everything with a single easy launch button.

3
🔌 Link your smart home

Tell Hestia where your home control system lives so she can see your lights, thermostats, and gadgets.

4
🧠 Give her a smart brain

Pick a thinking partner like Claude or GPT to help her understand and respond cleverly.

5
💬 Start chatting

Open the chat window, ask about your home or say 'turn on the kitchen lights,' and watch her respond live with thinking animations.

6
📈 Watch memories grow

See a beautiful 3D map of everything she learns about your home, rooms, and habits building over time.

🏠✨ Your home gets smarter

Hestia now remembers your preferences, runs voice commands perfectly, and makes your smart home feel truly alive and personal.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is HestiaClaw?

HestiaClaw is a JavaScript-powered, self-hosted AI assistant that plugs directly into your Home Assistant environment, letting you chat, query devices, and trigger automations via browser or voice. It remembers past interactions, learns about your home setup through a persistent knowledge graph, and actually does things like controlling lights or checking states without cloud dependencies. Docker Compose spins up the full stack – React frontend, Node backend, Neo4j graph DB – in one command.

Why is it gaining traction?

The killer hook is its layered memory: pinned facts for reliability, active recall for context, and a visual 3D graph that grows as it learns your home environment. Developers dig the MCP tool support for 87+ HA actions, runtime model switching (Claude, GPT, Ollama), and drop-in skills via simple markdown files. Voice ties into HA's Assist pipeline seamlessly, making it feel like a native extension.

Who should use this?

Home Assistant power users wanting local AI that evolves with their setup, without subscriptions or data leaks. Ideal for smart home tinkerers automating routines via chat, or HA satellite owners routing voice commands through a capable assistant. Skip if you're not running HA or prefer managed services.

Verdict

Grab it if Home Assistant is your hub – the memory that actually remembers and learns sets it apart for real-world use. 1.0% credibility score reflects 18 stars and early polish, but thorough docs and quickstart make it playable today; fork and contribute to mature it.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.