Rate-limitprotectionfromrunawayautomations

Automations misfire. A looping trigger, a bulk import, a sync that touches every row — suddenly something is firing hundreds of cache refreshes a minute. On a setup wired straight to Airtable, that's how a quiet Tuesday turns into a down site and an afternoon of detective work.

VibeKit assumes this will happen eventually and is built so that when it does, the failure lands in exactly the right place: the automation, not your site.

Eases off before it breaks

When refreshes from an automation, import, or sync start arriving too fast, VibeKit doesn't cut them off — it applies progressive back-pressure, slowing each refresh a little more than the last. A brief burst rides through untouched; a genuine flood gets gently squeezed instead of hammering anything downstream.

Fails the run, not your site

Past a threshold, VibeKit rejects the refresh with an error. That error fails the Airtable automation run — which is precisely what you want, because Airtable then flags the misbehaving automation by name. You find out which one went wrong without digging through logs or guessing which of your cache-refresh hooks was responsible.

Meanwhile, nothing happened to your visitors. Your site kept serving from cache the entire time. A runaway automation cannot take it down — the worst it can do is fail its own run.

A paper trail you don't have to watch

Every ease-off is recorded. It shows up in the admin's recent alerts, and if you subscribe to email digests, it's summarised in your inbox on the schedule you choose. When everything is quiet, you hear nothing.

Why this matters

The point of handing refresh controls to Airtable — buttons, formulas, automation scripts — is that editors keep your site fresh without touching a dashboard. That only works if a mistake on the Airtable side can't hurt the site. Flood protection is the guard rail that makes the rest of cache control safe to hand out freely.