Security quick test for AI applications

[DOPLNIŤ]

1/7 Which tool was the application built in?
2/7 Are API keys or secrets anywhere in the frontend or in the repository?
3/7 Does your database (e.g. Supabase) have row-level security (RLS) switched on?
4/7 Does the server verify sign-in and permissions on every request?
5/7 Do you have rate limiting on sign-in and the API?
6/7 Do you make backups and have you tried restoring them?
7/7 Who updates the application and watches for errors?