Monitor your API for free
Know when it goes down or gets slow. Test an endpoint right now — no account, no card, nothing to install.
HTTP and HTTPS, GET only. Don't put secrets in the URL.
What happens after the test
- Every 5 minutes, from the cloud, we ask your endpoint the same question you just did.
- Two failures in a row — not one — open an incident, so a blip does not wake you up.
- You get one email when it goes down and one when it comes back. Never a stream.
- Every day, we read your TLS certificate and warn you 14, 7, 3 and 1 days before it expires — the outage you can see coming, and the one nobody notices until it happens.
- Every Monday, a short report: uptime, number of checks, incidents, average and worst latency.
- A public status page, if you want one, with 30 days of history you can share or embed as a badge.
What "free" means here
- 5 monitors
- A check every 5 minutes
- 30 days of history
- Certificate expiry warnings
- Email alerts and weekly reports
- Public status pages and badges
No card, no trial, no seat count. If you need more, the things we are thinking about are listed in your dashboard and you can tell us which one matters — that is all it does.
Other checks you can run here
HTTP status code checker — what a URL answers, and what the code means.
SSL certificate checker — expiry date, issuer, and the names it covers.
Redirect checker — the whole chain, hop by hop, with timings.
CORS checker — the preflight a browser sends, and the rule that failed.
API response time test — three samples, so one slow request does not look like a trend.
Questions
Which addresses can I check?
Any public HTTP or HTTPS endpoint, on ports 80, 443, 8080 or 8443. Local and private addresses
(localhost, 10.x, 192.168.x, cloud metadata) are refused: checks run from
the internet, so an answer from them would be meaningless, and allowing them would make this tool a proxy into
other people's networks.
What counts as "up"?
By default, any answer with a 2xx or 3xx status; a 4xx or 5xx, a timeout, a DNS failure or a refused connection is a failed check. Redirects are followed, and the final address is shown. For a webhook receiver or an authenticated API — the kind that answers 401 or 405 to a plain GET — switch the monitor to reachable and any answer counts as up. That one switch is the difference between useful alerts and alerts you learn to ignore.
Do you read my responses?
No. Checks are GET requests and the body is never downloaded — only the status line and how long it took. We store the status code, the duration and the error, nothing else.
Will you email me things I did not ask for?
No. You get a sign-in code when you ask for one, alerts for monitors you created, and the Monday report — each switchable in your dashboard, and every message carries an unsubscribe link.