Guides
Security architecture guides for micro SaaS and AI/LLM platforms.
Every guide starts with the fix, then the reasoning, then a copy-pasteable snippet. Filter by the pattern or the stack you're running.
The one header check that leaks every tenant's data
A reporting endpoint trusts an X-Tenant-Id header instead of the verified session. Here's the exact request an attacker sends, and the two-line fix.
Why your AI agent shouldn't run with the company's service account
Excessive agency (OWASP LLM06) explained through a support-ticket agent that can be hijacked into reading every customer's data, not just the one it was asked about.
Auditing IAM roles before they become the incident
A short AWS CLI workflow for finding wildcard actions, unused roles and stale trust policies, plus how to prioritize what you find.
Want this checked on your system?