AI-generated doesn't have to mean insecure. Here is exactly what we do — for every app, every build, by default.
Each Kridona app deploys as its own Cloudflare Worker with its own dedicated database (Cloudflare D1). There are no shared tables, no shared runtime, no multi-tenant rows to leak across. A bug in one app cannot touch another app's data — including yours.
This is a structural guarantee, not a configuration flag. Category-wide incidents where thousands of generated apps shared a misconfigured database simply cannot happen here, because there is nothing shared to misconfigure.
Every build's code passes through an automated scan for dangerous patterns — data exfiltration to external servers, hardcoded credentials, injection primitives. The same scan gates the Community Store: no seller app can be listed for sale without passing security validation. You can see your own app's scan results in the Publish dialog.
Everything is served over HTTPS on Cloudflare's edge. Custom domains get TLS certificates provisioned automatically — there is no way to run a Kridona app without encryption in transit.
Your app's code is yours: export it as a ZIP or push it to GitHub at any time. No lock-in also means no hostage data — your app's database contents belong to you.
If you believe you've found a security issue in Kridona or in a generated app, email security@kridona.com. We read every report.