Articles

How software works underneath

Breakdowns of security, reliability and AI-written code, each with the file:line patterns behind it.

app-architecture.map3 layers · 6 questions
  1. Screen & bundleQ1 · WorksQ2 · Keys
    SettingsForm.tsx · NEXT_PUBLIC_*Optimistic UI lies and privileged keys in browser JS.
  2. Server & webhooksQ2 · AuthQ5 · Failure
    projects.ts:17 · stripe/route.ts:29Cookie-only checks, missing orgId gates and replayed events.
  3. Data & modulesQ3 · FastQ4 · ScaleQ6 · Maintain
    dashboard/page.tsx:22 · lib/utils.tsN+1 query loops, unbounded tables and tangled files.

All breakdowns

Patterns that AI-written code often gets wrong, each traced to the line and the fix.

Code review and hardening · 6 min read

How We Review AI-Built Codebases: Six Questions and File-Line Evidence

Automated scanners flag syntax patterns while missing broken authorization, silent failure paths, and client-trusted state. Here is how a Codeneath Review inspects an application from entry route to database row.

Security and vulnerabilities · 7 min read

Authentication Traps Under the Hood of AI-Generated Web Apps

Sign-in screens look finished long before authentication is safe. We break down cookie-presence checks, client-supplied user IDs, unverified webhooks, and token leakage.

Want this check on your code?

A Codeneath Review answers the six questions for your app, with a fix for every finding.