Open source · Read only · Self hosted

Ask your homelab anything.

Daemon is a local AI assistant that reads your infrastructure and answers questions in plain English. No dashboards. No agents. No setup hell. Runs on a Pi 4.

Get Daemon Cloud — $5/mo View on GitHub
daemon
$ daemon ask "why is my plex container using so much RAM" Your plex container is consuming 4.2GB RAM (87% of its limit). This is likely caused by transcoding. Consider setting a RAM transcode limit in Plex settings or raising the container cap.
$ daemon ask "is anything on my network that shouldn't be there" All detected devices match known MAC prefixes. Nothing unexpected. Network looks clean.
What it does

Everything you need. Nothing you don't.

Daemon is a single tool with a single job. Ask it about your homelab and get a real answer.

Ask anything
Plain English questions
Ask about containers, VMs, disk usage, network devices, resource hogs. Get answers that make sense.
Documentation
Auto-document your setup
Point Daemon at a container or VM and it writes the description for you. Export your entire lab to markdown in one command.
Memory
Sticky notes
Pin answers, set reminders, attach notes to specific resources. Get them emailed to you automatically.
Integrations
Local REST API
Every feature is exposed as a clean local API. Build Home Assistant cards, Discord bots, Grafana panels, anything you want.
Privacy
Read only. Always.
Daemon never executes commands on your system. It reads, reasons, and reports. Nothing else. Nothing phoning home.
AI
Your model, your key
Use OpenRouter for cloud models or run Ollama locally. Your key is stored locally in config.yaml and only used to reach your provider directly. It never leaves your machine.
How it works

Dead simple by design.

No agents. No execution. No moving parts. Just your homelab data plus a question equals a plain English answer.

01
You ask a question
In the browser UI or via the CLI. Plain English, no special syntax.
02
Daemon reads your lab
Live data pulled from Docker, Proxmox, TrueNAS via read-only API calls. Nothing is cached remotely.
03
AI reasons about it
Your question and your homelab state go to your chosen AI provider. The model reads the actual data and answers specifically.
04
You get an answer
Plain English. Specific to your setup. Real container names, real numbers, real suggestions.
Integrations

Works with what you already run.

Core integrations are built in. Everything else is a community PR away.

Docker Proxmox TrueNAS Unraid Home Assistant pfSense Ollama Portainer

Green means supported today. Grey means the API is there and the PR is welcome.

Pricing

Free forever. Cloud AI when you want it.

Daemon is open source and free to run locally with Ollama. If you want the easiest path to cloud AI without managing keys, add the Daemon Cloud subscription.

Self hosted
Freeforever
Open source. Run locally with Ollama or bring your own OpenRouter key.
  • Full source code on GitHub
  • Bring your own API key
  • Run models locally with Ollama
  • All integrations included
  • Community support
Get started on GitHub
Daemon Cloud
$5per month
No API keys to manage. One token, cloud AI included, cancel anytime.
  • Everything in Self hosted
  • Managed cloud AI included
  • No provider account needed
  • Sensible model selection
  • Email support