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.
What is 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.
Overview of home-generative-agent
What it is
Home Generative Agent (HGA) is a Home Assistant integration that brings a generative AI agent into your smart home. It provides conversational control over every HA entity, camera understanding with face recognition, long-term semantic memory, and the Sentinel anomaly engine. It is built on LangGraph and uses tools to control entities, query history, watch cameras, and write real HA automations. It supports cloud LLMs (OpenAI, Gemini, Anthropic) or fully local models via Ollama or any OpenAI-compatible server.
Key features
- Conversational control: talk to your home in natural language to turn things on, check status, ask questions.
- Automation creation: describe what you want in chat and the agent writes and registers the HA automation; a Critical Action PIN can hold automations that unlock doors or open garages for PIN confirmation.
- Camera & image analysis: ask the agent what it sees in any camera; proactive motion-triggered analysis with anomaly detection; works with Axis, Ring via ring-mqtt, Reolink, UniFi Protect, and any camera that exposes a motion entity or recording state.
- Sentinel anomaly detection: deterministic rules watch for security and safety issues (unlocked locks, open entries, unknown people) and alert your phone; includes a built-in Home Assistant security audit that flags admin accounts, access tokens, exposed ports, and more.
- Face recognition: identify people in camera frames and personalize alerts (requires the optional face-service).
- Long-term memory: semantic search over past conversations using pgvector; the agent remembers your preferences and context.
- Streaming responses: first tokens appear word-by-word in the HA conversation UI.
Use cases
- Controlling your smart home with natural language commands.
- Creating automations without writing YAML.
- Monitoring camera feeds with proactive anomaly alerts and face recognition.
- Receiving proactive security and safety alerts from the Sentinel anomaly engine.
- Running a fully local AI agent that never sends data outside your home.
Getting started
See the repository README for the full installation guide. The quick start requires Home Assistant OS or Supervised and involves installing the PostgreSQL with pgvector app, then configuring the integration.
License and pricing
MIT license. Open source.
Based on the project's public README and repository metadata.