Glossary

Web, database and AI security glossary

Short answer

Plain-language definitions of the security terms that appear in Hackaiz reports, so founders and developers can understand a finding without a security background.

By the Hackaiz team · HACKTUALIZ Inc. · Updated

Start my first scan

Terms

  • RLS (Row Level Security): database rules deciding which rows each user may read or change. Without them, a public key can read the whole table.
  • Service-role key: a Supabase key that bypasses all access rules. It must stay on the server, never in the browser.
  • SSRF (Server-Side Request Forgery): tricking a server into fetching an address chosen by the attacker, often an internal one.
  • CSP (Content-Security-Policy): a header telling the browser which scripts and resources a page may load, limiting injected code.
  • Clickjacking: hiding your page inside another site’s frame to trick users into clicking. Blocked by frame-ancestors or X-Frame-Options.
  • CORS: browser rules for which other sites may call your API. A wildcard with credentials is dangerous.
  • IDOR: accessing another user’s data by changing an ID in a request because ownership is not checked.
  • Prompt injection: text that makes an AI model ignore its instructions, reveal hidden context or misuse its tools.
  • Referrer-Policy: a header controlling how much of your URL is sent to other sites when users follow a link.
  • DMARC: an email record telling receivers what to do with mail that fails sender checks; p=none only monitors.