BackgroundApproximately half of 7-year-old children do not meet physical activity (PA) recommendations. Interventions targeting primary school children’s afterschool discretionary time could increase PA. Teaching assistants (TAs) are a school resource and could be trained to deliver after-school PA programmes. Building on earlier work, this paper describes the protocol for a cluster randomised feasibility study of a teaching assistant-led after-school intervention aimed at increasing PA levels of year 4 and 5 children (8–10 years old).MethodsPhase 1—pre-baseline: 12 schools will be recruited. In all schools, self-reported PA will be measured in all consenting year 3 and 4 children. In four schools, pupils will additionally wear a waist-worn Actigraph accelerometer for 7 days.Phase 2—baseline: schools will be randomised to one of two enhanced recruitment strategies being tested for children: (1) a club briefing and (2) the briefing plus a taster Action 3:30 session. Up to 30 children per school will be able to attend Action 3:30 sessions and will provide baseline data on height, weight, psychosocial variables and accelerometer-measured PA.Phase 3—intervention and follow-up: Schools randomised into intervention or control arm. Intervention schools (n = 6) will receive a 15-week after-school programme when children are in years 4 and 5, run by TAs who have attended a 25-h Action 3:30 training programme. Control schools (n = 6) will continue with normal practice. Follow-up measures will be a repeat of baseline measures at the end of the 15-week intervention.Phase 4—process evaluation: session attendance, perceived enjoyment and perceived exertion will be assessed during the intervention, as well as the economic impact on schools. Post-study qualitative assessments with TAs, school contacts and pupils will identify how the programme could be refined. Accelerometer-determined minutes of moderate-to-vigorous physical activity (MVPA) per day will be calculated as this is likely to be the primary outcome in a future definitive trial.DiscussionThe Action 3:30 cluster randomised feasibility trial will assess the public health potential of this intervention approach and provide the information necessary to progress to a definitive cluster randomised controlled trial.Trial registrationISRCTN34001941. Registered 01/12/2016.