A practical post on hardening Laravel projects against destructive AI agent behavior. It walks through the PocketOS incident and shows concrete safeguards like deny rules, credential isolation, and production guardrails.
PHP
I don’t write code anymore
Jason McCreary explains how he went from using AI for tedious snippets to letting agents handle nearly all coding, testing, and review work. It is a clear account of the guardrails, workflow shifts, and mindset Read more…