Objective Increasing effectiveness and sustainability of universal school‐based eating disorder prevention is needed. This study adapted two existing selective prevention programmes for universal delivery, investigating feasibility, acceptability, and preliminary effects when delivered by trained teachers to classes of mixed‐gender adolescents. Method A three‐arm controlled pilot study with Year 9 students (N = 288; Mage = 13.61 SDage = .50). Three schools in south Wales and south‐west England were allocated to mindfulness‐based intervention (MBI), dissonance‐based intervention (DBI), or classes‐as‐usual (CAU) control. Self‐reported eating disorder risk factors were collected at baseline, 6‐week post‐intervention, and 2‐month follow‐up. Focus groups were conducted post‐intervention. Results Delivery and evaluation were feasible, allowing for flexibility in scheduling, with good retention. Student and teacher ratings indicated moderate acceptability of both interventions, with recommendations for refinement. Mixed model analyses, controlling for baseline, showed significant effects of condition across post‐intervention and follow‐up for body esteem (DBI > CAU; Cohen's d = .34) and positive affect (MBI > CAU, d = .58). For girls only, both MBI and DBI improved body satisfaction and internalization, and the MBI additionally resulted in improved weight and shape concerns, negative affect, and life disengagement (d's = .39–1.12), across post‐intervention and follow‐up. Discussion Selective eating disorder prevention programmes based on cognitive dissonance and mindfulness can be delivered universally in schools, by teachers, allowing for appropriate flexibility necessary for real world implementation. Moderate acceptability indicates areas for improving content and delivery; positive effects on key outcomes are encouraging. These findings provide support for further robust evaluation. Public Significance Existing universal eating disorder prevention is limited by small effects and reliance on highly trained facilitators. This study is the first to adapt mindfulness‐ and dissonance‐based interventions for delivery by teachers, to adolescents of all genders in a classroom setting. Delivery was largely feasible and acceptable, and both interventions showed significant effects across key risk factors for eating disorders, with larger effect sizes than found previously. This underpins further robust evaluation.