Introduction: Stroke is characterized by an obstruction or reduction in blood supply, leading to decreased oxygen supply to the brain. This disease is the second leading cause of death worldwide and a leading cause of disability, including ankle disorders. Robot-Assisted Therapy (RAT) presents as a potentially effective ankle rehabilitation approach to improve walking and balance in stroke survivors. Unlike conventional ankle rehabilitation, RAT provides more repetitive and precise training. To address the gap in stroke rehabilitation, particularly in ankle function, this study was conducted to investigate the efficacy of various RATs in ankle rehabilitation in stroke patients. Methods: We conducted a literature search through PubMed, Scopus, Proquest, Sage, and EBSCOhost to assess RAT until August 12, 2024. The keywords used were (“Stroke” OR “Post-stroke” OR “Stroke rehabilitation”) AND “Robots” AND “Ankle” and their synonyms. Results: Five studies, with a total of 156 samples, were included in the qualitative and quantitative analyses. Meta-analysis showed that RAT ankle significantly improved walking speed with a Standardized Mean Difference (SMD) of 1.13 (95% CI: 0.32, 1.94; P = 0.006) and Active Range of Motion (AROM) with a SMD of 2.08 (95% CI: 1.5, 2.66; P < 0.00001). However, other parameters assessed in this study, namely the Berg Balance Scale (BBS) and Fugl-Meyer Assessment for Lower Extremities (FMA-LE), did not show significant effects in the intervention group compared to the control group. Conclusion: Robot-Assisted Therapy (RAT) shows promising potential to improve ankle rehabilitation in stroke patients. In addition, this study proves that RAT significantly improves walking speed and active ankle range of motion.