Abstract: Techniques for implementing smart annotation of digital content on a computing device are described in this document. In one embodiment, a method includes receiving an annotation to a document displayed in an application on a computing device. The application is in an annotation mode in which any received user input is recognized as annotations but not a part of the underlying content. The method also includes determining whether the annotation is related to an editing mark applicable to the underlying content of the document. In response to determining that the received annotation is related to an editing mark, determining an editing operation corresponding to the editing mark and performing the editing operation to the underlying content in the displayed document without exiting the annotation mode of the application.