Your intent, in every AI call - every model means what you meant, not what it guessed.
Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts — it cannot prove the absence of malicious code.
Your intent, in every AI call - every model means what you meant, not what it guessed.