The offer
Free. No card. No trial timer.
There is no paid tier, because billing has not been built. That is the honest reason, and it is better than inventing a “Pro” column with a Contact Us button behind it.
| Included | |
|---|---|
| Monitors | 20 per workspace |
| Monitor types | All seven — HTTP, TCP, heartbeat, certificate, domain, DNS, email auth |
| Check interval | From 20 seconds |
| Alert channels | Email, webhook, ntfy, Telegram — any number per monitor |
| Reminders while down | Included |
| Second opinion | Included |
| Check history | 7 days |
| Team members | Any number, in one workspace |
| Cost | Nothing |
What you should know before relying on it
Stated here rather than in a footnote, because you are deciding whether to trust this with something that matters.
It is beta, and there is no SLA. No uptime guarantee, no support commitment, no compensation if a check is missed. If your monitoring must itself be contractually reliable, you need a vendor who will sign something, and this is not that yet.
One probe location. Not multi-region.
20 monitors per workspace. Not a pricing lever — a deliberate cap while the system is young.
7 days of check history. Raw results are deleted after 7 days and there is no long-term rollup, so uptime percentages cover the last 24 hours and the last 7 days. Nothing longer.
No status pages, maintenance windows, incident acknowledgement or on-call scheduling. All designed, none built — see the roadmap.
Will it stay free?
The parts you are using now — the seven monitor types, the alert channels, the reminders, the second opinion, at the current limits — will stay free.
If paid tiers arrive, they will be for things that cost real money to run: more monitors, longer retention, higher check frequency. Nobody will wake up to find their existing monitors switched off, and there will be notice.
That is a statement of intent from a solo developer, not a contract. Weigh it accordingly.
Why free at all?
Because a monitoring product with no users has no signal. Every improvement so far has come from running it against real targets and finding out where it was wrong — a DNS response shape that made a dead domain look healthy, a parsing bug that froze the prober, a seed helper that had been failing silently for days.
Charging before that loop has run would be charging for something that hasn’t been tested by anyone but its author.