随着生成式AI技术的突破性发展，其在国际中文教育中的应用日益广泛，为解决教学资源不均、真实语伴匮乏等长期难题提供了新思路。但生成式AI的输出质量高度依赖提示词，提示策略作为人机交互的核心接口，其科学应用成为提升教学实效的关键。本研究以国际中文教育场景为依托，结合图式理论、社会学习理论等教育学与心理学理论，提炼出零样本、少样本、思维链、教学目标锚定四种适用于教学场景的提示策略，通过案例分析，明确各策略在典型教学环节的实践方式，同时从教师、学生、课程资源三个层面提出针对性应用建议，并梳理相关应用局限与未来优化方向。科学运用这些提示策略，可充分发挥生成式AI的辅助价值，助力教师提升教学效率，推动国际中文教育高质量发展。

With the breakthrough development of generative AI technology, its application in international Chinese language education has become increasingly widespread, providing new solutions to long-standing problems such as uneven distribution of teaching resources and a lack of real-language partners. However, the output quality of generative AI is highly dependent on prompts. As the core interface of human-computer interaction, the scientific application of prompt strategies has become the key to improving teaching effectiveness. Based on the scenarios of international Chinese language education, this study combines pedagogical and psychological theories including Schema Theory and Social Learning Theory, and extracts four prompt strategies suitable for teaching scenarios: zero-shot, few-shot, chain-of-thought, and teaching goal anchoring. Through case analysis, it clarifies the practical application of each strategy in typical teaching links. Meanwhile, it puts forward targeted application suggestions from three dimensions: teachers, students, and curriculum resources, and sorts out the limitations and future optimization directions of relevant applications. The scientific application of these prompt strategies can give full play to the auxiliary value of generative AI, help teachers improve teaching efficiency, and promote the high-quality development of international Chinese language education.