About SensaCat
SensaCat is a monitoring platform for the failures that leave every server green: a cron job that quietly stopped running, a webhook that never arrived, a business process that stalled halfway, an API credential about to expire over a weekend.
Why it exists
Infrastructure monitoring answers one question well: is the machine up? It is very bad at the question that actually costs money, which is whether the work got done.
A payment can be captured while the fulfilment webhook never fires. A nightly export can be switched off by a deploy and nobody finds out until a customer asks for a report. A certificate expires on a Saturday. In each case the dashboards stay green, the error rate stays at zero, and the first alert is an email from a customer.
SensaCat exists to make silence itself the alert. You describe what should happen and how long it may take, and you hear about it when that stops being true.
What it watches
- Heartbeat monitors — cron jobs, queue workers and scheduled tasks check in on an interval. Miss the window and the job shows as late; miss the grace period too and it raises an alert.
- Business flows — multi-step processes such as checkout to fulfilment or sign-up to onboarding. Each step has a maximum gap, and an instance that outlives its next step's gap is reported as stalled.
- Credentials — API keys, tokens and certificates, with reminders that start as warnings and turn critical as the expiry date approaches.
- Domains — registration expiry, SSL certificate expiry, and uptime probing with two consecutive failures required before anything is raised.
Alerts fan out to email, Slack, Discord, SMS and phone calls, and escalate when nobody acknowledges them.
How it is built
A few decisions we are not planning to reverse:
- One alert per incident. A stalled flow or a silent job raises a single alert when it breaks, not a fresh one on every check. Monitoring you learn to ignore is not monitoring.
- Nothing to install. No agent and no SDK. Sending an event is an HTTP POST, authenticated with the resource's own ingest token and your account API key, so a leaked URL on its own cannot write data.
- Metadata only. An event carries timestamps, the flow or monitor it belongs to, the id of the record moving through it and a status. We do not ask for payload contents and do not store them.
- Late is not down. A job that runs a little behind schedule is shown as late on the dashboard and wakes nobody. A 4xx from a domain is treated as up, because bot protection refusing a checker is not an outage.
Who runs it
SensaCat is a small, independently funded product. It is not venture-backed, it has no sales team, and the person who answers support is the person who writes the code.
That shapes what you can expect: the pricing page shows the real limits rather than "contact us", the free plan is a real plan rather than a trial, and a feature that is not built yet is not listed as if it were.
Talk to us
Questions about the product, pricing, security or anything else go to [email protected], or through the contact form. Bug reports and feature requests are welcome at the same address.
If you want to try it, the pricing is public and 20 free tools need no account at all.