Recipes now have better support for exit codes. If a command throws an error OR returns a non-zero exit code, the recipe will stop and the exit code of the last command will be returned as the exit code from the recipe command. And if it was a non-interactive shell, the exit code will flow all the way back to the operating system from the box binary. Also, running "exit" inside of a recipe will no longer exit the entire shell, but just that recipe execution. This give you a lot better control over your recipes.