Huginn hosting
Agents that watch the web and act when something changes.
Huginn hosting at AIHostBox means a dedicated, single-tenant Huginn instance on our infrastructure: we install it, pin it to a reviewed release, put it behind TLS on your domain, back it up nightly and keep it patched. Plans start at $5.99 per month and the admin account, the data and any API keys are yours.
Huginn is a long-running, MIT-licensed Ruby on Rails project, started by Andrew Cantino and kept going by community contributors, for building agents that watch and act on your behalf. Each agent does one small job: a WebsiteAgent scrapes a page with CSS or XPath selectors, an RssAgent reads a feed, a TriggerAgent matches a pattern, an EmailDigestAgent bundles results. Agents pass JSON events to each other, Liquid templates reshape them, and groups of agents are saved as scenarios you can export and share.
It differs from newer workflow builders in its unit of work. Instead of a flow that runs once per trigger, Huginn keeps dozens of agents polling on their own schedules and storing the events they produce, closer to a personal IFTTT that remembers history. That is why it needs a server: the scheduler and background workers must run continuously for checks to fire on time. The codebase now also includes an OpenAI LLM agent that can call any OpenAI-compatible endpoint, Ollama included.
Huginn hosting plans
Every plan is a dedicated single-tenant instance with NVMe storage, nightly off-box backups, TLS and managed patching. Quarterly, semi-annual and annual terms are discounted at checkout. Model tokens are not included — you bring your own key.
How we run Huginn
The official Huginn container from ghcr.io/huginn/huginn (Rails web server, scheduler and delayed_job workers) connected to a separate PostgreSQL database with DATABASE_ADAPTER=postgresql.
What we configure for you
- APP_SECRET_TOKEN generated fresh instead of the example value
- DOMAIN set to your hostname with FORCE_SSL=true
- INVITATION_CODE so only people you invite can register
- SMTP_SERVER, SMTP_USER_NAME and EMAIL_FROM_ADDRESS so email and digest agents can send
- TIMEZONE so schedules fire at your local hours
- OPENAI_API_KEY and OPENAI_BASE_URL if you use the LLM agents
Which plan fits
Starter (2 GB, 1 vCPU) runs a personal set of watchers comfortably, typically dozens of agents checking every few minutes to hourly. Plus (4 GB, 2 vCPU) suits heavier scraping, several users or agents that fire frequently. Pro (8 GB, 3 vCPU) is for large event volumes where the database and workers both need headroom.
The honest limits
Huginn is volunteer-maintained and moves slowly; some agents target services whose APIs have changed or closed, so check an agent still works before building on it. The WebsiteAgent does not execute JavaScript, so script-heavy pages need an external rendering service.
What people run Huginn for
Change detection
Know when a competitor’s pricing page changes.
Personal feeds
Compose sources into one digest.
LLM step in a monitoring chain
Summarise or classify new items with the OpenAI LLM agent against a hosted model or your own Ollama.
Webhook glue
Receive events with a WebhookAgent and forward reshaped JSON with a PostAgent.
Licence and version policy
| Item | Position |
|---|---|
| Upstream licence | MIT |
| Sold as managed hosting | Yes |
| Tenancy | Single tenant — one customer per instance |
| Patching | Critical advisories within 24 hours, high severity within 72 |
Our full position on licences, including the applications we decline to sell, is on the licensing page.
Works well with Huginn
More Automation we host
Huginn hosting: your questions
Can Huginn use an LLM?
Which database does Huginn use here?
Can I import scenarios from another Huginn install?
How do I stop strangers signing up?
Are shell command agents allowed?
Run Huginn without running a server
A dedicated Huginn instance, deployed and maintained by us. Month to month, cancel whenever.