Alcohol and cannabis are the first and second most used substances among adolescents. Adolescence is a period of considerable development, making the adolescent brain particularly vulnerable to negative effects of alcohol and cannabis use. Developing and testing interventions that target both alcohol and cannabis use during adolescence are vital to decreasing costly consequences. Biases in cognitive processing of drug-related stimuli play an important role in the development and maintenance of problematic substance use. The Approach-Avoidance Task (AAT) is a computerized program, effective in assessing implicit approach biases for both alcohol and cannabis, in which participants make approach or avoidance movements in response to an irrelevant feature of an image presented on a screen (e.g., push when in portrait, pull when in landscape). A modified version of the AAT is also used as an approach bias modification (ApBM) intervention, to retrain participants' implicit biases toward or away from stimuli by presenting the target stimuli predominantly in one format (e.g., push or pull). Despite research demonstrating the effectiveness of AAT interventions to reduce problematic alcohol and cannabis use, there is a dearth of research examining this intervention among adolescents. This protocol paper describes a NIDA-funded randomized control trial (RCT) to evaluate an integrated mobile ApBM intervention to target co-occurring alcohol and cannabis use among treatment-seeking adolescents. Outcomes will be measured from pre-treatment through a three-month follow-up. The sampling procedures, assessment protocol, description of the intervention, and planned statistical approaches to evaluating outcomes are detailed. Clinical and research implications of this work are also discussed.