security-hook-blocks-literal-exec-eval-in-prose

The security_reminder_hook regex fires on literal exec( and eval( tokens anywhere in file content — including prose descriptions, comments, and handoff docs that describe the hook itself. This causes false-positive blocks when writing skill files or handoff documents that document the hook’s behavior. Workaround: use descriptive language (‘shell execution pattern’ instead of ‘exec(’) to avoid triggering the regex.