New: AI Script Generator

RefreshPilot · Use case

Auto refresh for uptime & error watch

Not everything deserves a paid monitoring service. For the admin panel, the staging site, the client's old website — a browser tab that reloads, checks for error words, and messages Slack is monitoring enough.

Disponibilité & surveillance d'erreurs

A starting configuration

Interval

Every 1–5 minutes, hard refresh on

Monitor

Keywords on error text; alert if the health line disappears

Alert

Webhook to Slack/PagerDuty + notification

Every website is a little different — start with this and adjust until it feels right.

How to set it up

1

Open what outside services can't see

The logged-in admin panel, the staging site, the checkout's last screen — RefreshPilot watches whatever your browser can see.

2

Hard refresh on a steady beat

Every 1–5 minutes, cache off — so a saved copy can't show you "everything's fine" while things burn.

3

Check the page for trouble

Alert when "500", "error" or "connection failed" appears — or when a healthy sign like "queue: ok" disappears. Missing words are as watchable as new ones.

4

Tell the team, not just yourself

Webhook into Slack or PagerDuty with what matched. Notification for whoever's on duty.

01 — The problem

The gap between "no monitoring" and "real monitoring"

Proper monitoring takes tools, setup and budget — so internal tools, test sites and small client sites usually get none, and failures announce themselves through an annoyed human. Uptime services only check public pages from outside; they can't see the error banner on your logged-in admin page.

Features doing the heavy lifting

Actualisation forcée

Actualisation forcée

Ignorez le cache et récupérez chaque octet directement depuis le serveur à chaque passage — pour les pages qui s'obstinent à vous montrer les données d'hier.

Veille de mots-clés

Veille de mots-clés

Confiez-lui les phrases qui décident de votre journée — « de retour en stock », « rendez-vous disponibles », un prix — et il vous prévient à l'instant où l'une d'elles apparaît sur la page, ou en disparaît.

Intégration webhook

Intégration webhook

Chaque détection peut appeler Slack, Discord, Telegram, ntfy ou un endpoint à vous — les changements de page deviennent des événements dans la stack que vous utilisez.

FAQ

Good questions

Des réponses claires sur le fonctionnement, ce qui est gratuit et ce que PRO débloque. Il manque quelque chose ? Nous répondons sous un jour ouvré.

Encore des questions ?

Dites-nous ce que vous surveillez — nous répondons à chaque message sous un jour ouvré.

Contacter le support
Does this replace real monitoring?

For big production systems, no — use proper monitoring there. This is for everything underneath: staging, internal tools, status pages, a client's WordPress. The monitoring that otherwise wouldn't exist at all.

What can it see that UptimeRobot can't?

Everything behind your login: admin panels, private staging sites, pages that load fine but say "database connection failed" in the middle. Outside services check if the page answers; this checks what the page actually says.

Can it alert when something disappears?

Yes — and watching for a missing word is often the sharper signal: a "last sync: ok" line vanishing, a timestamp going stale. Alert-on-disappear is built in.

What if the computer sleeps or the browser restarts?

A sleeping computer pauses any in-browser tool — for round-the-clock coverage, use a machine that stays awake. Auto-start rules re-arm every watch when the browser opens, so a restart doesn't silently kill your monitoring.

Démarrer

Cessez de surveiller la marmite.
Nous sifflerons quand ça bout.

Installez RefreshPilot, pointez-le sur n'importe quelle page et passez à autre chose — votre premier moniteur est en ligne en une minute environ.

Forfait gratuit à vie · Aucune carte requise · Premier moniteur actif en une minute