The technical debt (TD) metaphor is widely used to encapsulate numerous software quality problems. She describes the trade-off between the short term benefit of taking a shortcut during the design or implementation phase of a software product (for example, in order to meet a deadline) and the long term consequences of taking said shortcut, which may affect the quality of the software product. TDs must be managed to guarantee the software quality and also reduce its maintenance and evolution costs. However, the tools for TD detection usually provide results only considering the files perspective (class and methods), that is not usual during the project management. In this work, a technique is proposed to identify/visualize TD on a new perspective: software features. The proposed technique adopts Mining Software Repository (MRS) tools to identify the software features and after the technical debts that affect these features. Additionally, we also proposed an approach to support maintenance tasks guided by TD visualization at the feature level aiming to evaluate its applicability on real software projects. The results indicate that the approach can be useful to decrease the existent TDs, as well as avoid the introduction of new TDs.