Hyperping

Promptfree, for everyone, and the only version there is

Open the prompt in Claude CodeOpen the prompt in CodexOpen the prompt in Cursor

Copy

Build me uptime checks and a status page, hosted somewhere my own infrastructure cannot take down with it.

STACK
- A Node checker running on a small VPS at a different provider from the thing it watches
- SQLite for results, a static status page published to object storage behind a CDN

CHECKS
- HTTP with an expected status and an optional body match, plus TCP and ICMP
- Every 60 seconds, from at least two regions; a check fails only when both regions agree
- Timeout of 10 seconds; record latency on every run, not only on failure

INCIDENTS
- Open one after three consecutive failures, close after two consecutive successes — never on a single sample
- An incident is a series of append-only updates: investigating, identified, monitoring, resolved
- Post-mortem field left empty until written

THE PAGE
- Rendered to static HTML on every state change and pushed to object storage — it must not depend on your app, your database or your network
- 90 days of daily uptime bars per component, computed from the check log
- A JSON endpoint with the same data

NOTIFICATIONS
- Email through a provider, plus webhooks; subscribers confirm by double opt-in
- Never send a notification for a flap that resolved inside one check interval

HONESTY
- Uptime is computed from checks, never entered by hand, and the page says so

Include a README on where to host the checker, and why it must not be the same provider.