Last released Aug 23, 2026
The trust layer for AI-written code: say what must never break, prove the protection holds, enforce it on every agent edit.
Supported by
