This paper presents an approach for the recommendation of pedagogical actions in Virtual Learning Environment (VLE), using automatic planning. The actions considered by the planner are structured by Bloom's Digital Taxonomy. The state is specified according to the student model. This model contains information provided by the VLE and the ASSIST questionnaire. A base of cases containing plans associated with student models is maintained and initially consulted in order to optimize the process. A proof of concept shows the feasibility of using Automated Planning effectively in the context of recommendations for generic pedagogical actions.