Emerging micropollutants, such as pharmaceuticals and microplastics (MPs), have become a pressing water environmental concern. The aim of this study is to synthesize chitosan sponges using graphene oxide (GO) and genipin (GP) for the removal of pharmaceuticals (diclofenac (DCF) and triclosan (TCS)) and MPs, verify their adsorption mechanisms, evaluate the effects of temperature, pH, and salinity on their adsorption capacities, and determine their reusability. The GO5/CS/GP sponge exhibited a macroporous nature (porosity = 95%, density = 32.6 mg/cm3). GO and cross-linker GP enhanced the adsorption of DCF, TCS, and polystyrene (PS) MPs onto the CS sponges. The adsorption of DCF, TCS, and PS MPs involved multiple steps: surface diffusion and pore diffusion of the sponge. The adsorption isotherms demonstrated that Langmuir model was the most fitted well model to explain adsorption of TCS (qm = 7.08 mg/g) and PS MPs (qm = 7.42 mg/g) on GO5/CS/GP sponge, while Freundlich model suited for DCF adsorption (qm = 48.58 mg/g). DCF adsorption was thermodynamically spontaneous and endothermic; however, the adsorption of TCS and PS MPs was exothermic (283–313 K). The optimal pH was 5.5–7 due to the surface charge of the GO5/CS/GP sponge (pHzpc = 5.76) and ionization of DCF, TCS, and PS MPs. As the salinity increased, DCF removal efficiency drastically decreased due to the weakening of electrostatic interactions; however, TCS removal efficiency remained stable because TCS adsorption was mainly caused by hydrophobic and π–π interactions rather than electrostatic interaction. The removal of PS MPs was enhanced by the electrostatic screening effects of high Na+ ions. PS nanoplastics (average size = 26 nm) were removed by the GO5/CS/GP sponge at a rate of 73.0%, which was better than that of PS MPs (41.5%). In addition, the GO5/CS/GP sponge could be recycled over five adsorption–desorption cycles.