Software testing is a quality activity with the purpose of revealing errors not yet discovered. Also, this activity verifies if the software realizes its functionalities conformance with its specification. As software testing activity evolves during software development, a large set of test cases can be generated, compromising the effort during the regression testing, for instance. Regression testing is a type of software testing to confirm that a new program or code change has not adversely affected existing features of the software. In this activity, the tester may not have enough time to run all test cases, needing to decide which test cases are best in terms of effectiveness in reveal faults. In another way, during code refactoring activities, code smells can be identified, which represents unwanted situations that can negatively impact software quality. This project proposes an approach that employs information about code smells to guide software testing activity. Our hypothesis is that the code smells can guide, for instance, the prioritization of test cases, focusing on effort reduction of software testing. I would like to acknowledged my advisor, Simone R. S. Souza for guiding and supporting in this work.