Learnings of integrating OpenClaw & Local LLM into my Home Assistant instance

Explore integrating OpenClaw and local LLMs with Home Assistant for Telegram control. Learn about agent management, LLM settings, and what to expect from this agentic AI setup.

Home Assistant OpenClaw Docker MCP Gemma 4 E4B

Overview

Does OpenClaw elevate my smart home experience on top of the already powerful Home Assistant?

By leveraging the OpenClaw add-on in Home Assistant and connect it to its MCP Server, I managed to control part of my smart home via Telegram. I explored the capabilities of the latest local LLMs (or “SLMs”?), such as Gemma 4 E4B, Qwen 3.5 9B, GPT-OSS-20B etc., and experienced some hit-and-misses along the way. I will share what worked well, what worked, and what didn’t.

Links

https://github.com/amandalam/Home-Assistant-Snippets (OpenClaw rela...

Tech stack

Home Assistant

Open-source home automation: Local control, privacy-first, integrating over 3,000 devices and services.

Home Assistant is the premier open-source platform for smart home control, prioritizing local processing and user privacy. It unifies over 3,000 distinct devices and services (Zigbee, Z-Wave, Matter, etc.) into a single, powerful system. Run the core on a Raspberry Pi or local server: Configure advanced automations, build custom dashboards, and utilize the Assist voice assistant for fast, on-premises control.

https://www.home-assistant.io/

View projects
OpenClaw

OpenClaw is the viral, open-source, autonomous AI agent: a self-hosted 'digital employee' that executes real-world tasks across your local machine and messaging platforms 24/7.

This is the next-generation autonomous AI agent, built by Peter Steinberger (founder of PSPDFKit). OpenClaw functions as a proactive, self-hosted assistant, running as a long-running Node.js service on your own hardware (e.g., a Mac Mini or VPS) for about $3–$5 per month. It integrates directly with chat apps (WhatsApp, Telegram, Discord) to receive instructions and report completions. The agent utilizes over 100 AgentSkills to execute complex, real-world workflows: clearing your inbox, writing code, managing documents, and checking you in for flights. The open-source project’s velocity is undeniable, having surpassed 100,000 GitHub stars quickly and reportedly driving a surge in Mac Mini sales.

https://openclaw.ai

View projects
Docker

Docker is the open-source platform that packages applications and dependencies into standardized, portable containers for consistent execution across any environment.

Docker is the industry-standard containerization platform, enabling developers to build, ship, and run applications efficiently. It uses the Docker Engine (the core runtime) to create lightweight, isolated environments called containers: these units bundle an application’s code, libraries, and configuration. This self-contained approach guarantees consistency, eliminating the 'it works on my machine' problem across development, testing, and production environments (local workstations, cloud, or on-premises). Docker debuted in 2013 and now serves over 20 million developers monthly, simplifying complex workflows like CI/CD and microservices architecture by leveraging tools like Docker Hub for image sharing and Docker Compose for multi-container applications.

https://www.docker.com

View projects
MCP

MCP is the open-source standard for securely connecting AI agents (like LLMs) to external tools, data, and enterprise workflows.

The Model Context Protocol (MCP) functions as a standardized integration layer: think of it as a USB-C port for AI applications. Developed and open-sourced by Anthropic, this protocol allows large language models (LLMs) to access real-time context and execute actions via external tools like GitHub, Jira, or proprietary databases . It uses a simple JSON-RPC interface to define tools, schemas, and endpoints, which enables AI agents to perform complex, state-changing tasks—such as creating a GitHub issue or running a test script—rather than just generating text . MCP is essential for building agentic AI systems that can autonomously pursue goals and operate within defined safety and permission boundaries .

https://modelcontextprotocol.io/

View projects
Gemma 4 E4B

Gemma 4 E4B is a high-efficiency 4-billion parameter model optimized for edge deployment and low-latency inference.

Google's Gemma 4 E4B leverages a refined 4B parameter architecture to bridge the gap between mobile hardware constraints and enterprise-grade reasoning. Built on the same technical foundations as Gemini, this model utilizes advanced quantization techniques to maintain 92% of full-precision performance while operating within a 2.5GB memory footprint. It excels in local RAG (Retrieval-Augmented Generation) workflows and structured data extraction, providing developers a private, offline-capable alternative to cloud-based APIs without sacrificing execution speed.

https://ai.google.dev/gemma

View projects

Learnings of integrating OpenClaw & Local LLM into my Home Assistant instance

Related talks