To identify and synthesise the content of knee bracing interventions in randomised controlled trials (RCTs) of knee osteoarthritis (OA). In this scoping review, three electronic databases (PubMed, Web of Science, Cochrane) were searched up to 10th June 2024. Nineteen previous systematic reviews of knee bracing for knee OA and four recent international clinical practice guidelines were also hand searched. Identified studies were screened for eligibility by two independent reviewers. Information on bracing interventions was extracted from included RCT reports, informed by Template for Intervention Description and Replication (TIDieR) guidelines. Data were synthesised narratively. Thirty-one RCTs testing 47 different bracing interventions were included. Braces were broadly grouped as valgus/varus, patellofemoral, sleeve, neutral hinged, or control/placebo knee braces. Brace manufacturer and models varied, as did amount of recommended brace use. Only three interventions specifically targeted brace adherence. Information on brace providers, setting, number of treatment sessions, and intervention modification over time was poorly reported. Adherence to brace use was described for 32 (68%) interventions, most commonly via self-report. Several mechanisms of action for knee braces were proposed, broadly grouped as biomechanical, neuromuscular, and psychological. Many different knee brace interventions have been tested for knee OA, with several proposed mechanisms of action, a lack of focus on adherence, and a lack of full reporting. These issues may be contributing to the heterogeneous findings and inconsistent guideline recommendations about the clinical effectiveness of knee bracing for knee OA to date.