goruck/home-generative-agent

AI agent for Home Assistant — talk to your home, create automations in plain language, analyze cameras with face recognition, and get proactive anomaly alerts. Cloud LLMs or fully local.

What it solves

Home Generative Agent (HGA) transforms a standard Home Assistant setup into an intelligent, conversational smart home. It solves the problem of rigid, manual automation by allowing users to control devices, create automations, and query home history using natural language, while providing proactive security monitoring.

How it works

HGA is a LangGraph-based agent that integrates directly into Home Assistant. It uses a combination of tools to interact with your smart home entities, a PostgreSQL database with pgvector for long-term semantic memory, and a "Sentinel" engine for deterministic security and anomaly detection. It supports both cloud-based LLMs (OpenAI, Gemini, Anthropic) and local models (via Ollama or OpenAI-compatible servers), ensuring that users can choose whether their data stays entirely local or uses cloud processing.

Who it’s for

This project is for Home Assistant users who want a sophisticated, AI-driven interface for their home automation. It is ideal for users comfortable with self-hosting (or using local LLM infrastructure) who want to move beyond simple voice commands to a system that can "reason" about home state, write automations on the fly, and perform security audits.

Highlights

  • Conversational Control: Talk to your home to manage devices, check status, or ask questions.
  • Automated Logic: Describe an automation in plain English, and the agent writes, registers, and manages the YAML code for you.
  • Proactive Security: The "Sentinel" engine performs automated security audits of your network and home devices, flagging vulnerabilities like exposed ports or unauthorized access.
  • Multimodal Capabilities: Analyze camera footage for object or face recognition and receive proactive alerts.
  • Privacy-First Options: Fully compatible with local LLMs, speech-to-text, and text-to-speech engines, keeping your data within your home network.

Related

  • Project
    Fullive-AI/Anima

    An open-source Agent OS for intelligent hardware that transforms smart home devices into proactive AI agents using a brain-skill-memory architecture.

  • Project
    jekalmin/extended_openai_conversation

    A custom Home Assistant component that extends OpenAI Conversation with function calling to allow an AI assistant to control devices, create automations, and fetch external data.

  • Project
    cqzyys/lang-agent

    A visual configuration platform based on LangGraph that allows users to build complex AI agents using a drag-and-drop interface with customizable state variables for precise logic control.

  • Project
    judahpaul16/gpt-home

    A self-hosted AI home assistant for Raspberry Pi that uses LiteLLM and LangGraph to integrate LLMs with home automation and personal productivity tools.

  • Project
    zhayujie/CowAgent

    CowAgent is an open‑source, self‑hosted AI assistant that plans tasks, runs tools, keeps a personal knowledge base, and can operate as a team of cooperating agents. It supports many LLM providers, multimodal input, dozens of chat channels (Telegram, Slack, WeChat, etc.), a plug‑in skill system, and a three‑tier memory with nightly “Deep Dream” distillation. Install via a one‑line script or Docker, manage everything through a web console, and extend the system through the public Skill Hub or custom tool integrations.