Porous β-CD polymer (CDP) is a good adsorbent with broad spectrum for removal of micropollutants, preparation of composites of CDP with magnetic materials is highly expected. In this work, a composite of magnetic graphene and CDP (magGO-CDP) was prepared, it was fully characterized and was used as an adsorbent to remove model pollutants, including methylene blue (MB), rhodamine B (RhB), fleroxacin (Fler), enrofloxacin (Enr), bisphenol A (BPA), bisphenol S (BPS), nabumetone (NaB) and indomethacin (Ind) in wastewater. Factors influencing the removal efficiency were studied. The removal rates of the eight pollutants can reach 101 ± 0.7 %, 94.7 ± 1.8 %, 83.3 ± 1.1 %, 91.9 ± 0.2 %, 58.9 ± 0.9 %, 77.4 ± 3.5 %, 96.2 ± 0.3 % and 94.7 ± 0.8 %, respectively. Their maximum adsorption capacities are 133, 178, 185, 34.8, 362, 247, 186 and 187 mg/g, respectively. Findings from kinetic studies suggest that the pseudo-second-order model can successfully describe the overall adsorption process. The adsorbent has good capability to resist fouling of natural organic matter, and it has good reusability and can be recycled for at least five times. It has a higher adsorption capacity than that of previously reported materials and could achieve good removal efficiency with smaller adsorbent dosage and shorter time. It was also used for simultaneous removal of multiple pollutants and removal of pollutants in simulated wastewater with satisfactory results. The magGO-CDP can serve as a promising candidate for wastewater remediation. The strategy for preparing the composite of CDP and magGO can be straightly extended to other matrixes to produce high performance adsorbents. [Display omitted] • A composite of magnetic graphene/β-cyclodextrin polymer (magGO-CDP) was prepared. • magGO-CDP had high adsorption capacity for pollutants up to 362 mg bisphenol A/g. • Removal efficiency of > 90 % can be achieved with 0.5 g/L adsorbent dosage in 1 min. • Simultaneous removal of multiple pollutants and simulated wastewater remediation. • The strategy can be extended to prepare composites of CDP and other matrixes. [ABSTRACT FROM AUTHOR]