🔧 I type-check AI-generated SDK code against the real package. Claude refused a third of my Stripe tasks.

I build a small tool called SDKProof. It measures whether AI coding agents write a library's current API or an older one they remember. A model solves 10-15 real tasks, each answer gets dropped into a project with the real installed package, then tsc --noEmit. Pass = compiles clean. No LLM judging another LLM, the compiler decides.

Last night I...