Abstract Background Globally, methamphetamine use has increased in prevalence in recent years. In Australia, there has been a dramatic increase in numbers of people seeking treatment, including residential rehabilitation, for methamphetamine use disorder (MUD). While residential rehabilitation is more effective for MUD than withdrawal treatment (i.e. “detoxification”) alone, relapse rates remain high, with approximately half of rehabilitation clients using methamphetamine within 3 months of rehabilitation. “Approach bias modification” (ABM) is a computerised cognitive training approach that aims to dampen automatically triggered impulses to approach drugs and drug-related stimuli. ABM has been demonstrated to reduce alcohol relapse rates, but no randomised controlled trials of ABM for MUD have yet been conducted. We aim to test whether a novel “personalised” form of ABM, delivered during rehabilitation, reduces post-treatment methamphetamine use, relative to a sham-training control condition. Secondary outcomes will include dependence symptoms, cravings, and approach bias. Methods We aim to recruit 100 participants attending residential rehabilitation for MUD at 3 sites in the Melbourne metropolitan area. Participants will complete baseline measures of methamphetamine use, craving, dependence severity, and approach bias before being randomised to receiving 6 sessions of ABM or “sham” training. In the active condition, ABM will be personalised for each participant, using those methamphetamine images that they rate as most relevant to their recent methods of methamphetamine use as “avoidance” images and using positive images representing their goals or healthy sources of pleasure as “approach” images. Approach bias and craving will be re-assessed following completion of training, and methamphetamine use, dependence, and craving will be assessed 4 weeks and 3 months following discharge from residential treatment. Discussion This study is the first randomised controlled trial of ABM for MUD and also the first ABM study to test using a personalised set of both approach and avoid images for ABM training. If effective, the low cost and easy implementation of ABM means it could be widely implemented as a standard part of MUD treatment. Trial registration Australian New Zealand Clinical Trials Registry ACTRN12620000072910. Registered on 30 January 2020 (prospectively registered): https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=378804&isReview=true