For AI agents to effectively execute multi-step operations and achieve complex objectives, they must be capable of planning their actions and managing tasks. This chapter concentrates on the prompt engineering techniques that make these agent capabilities possible. You will learn to design prompts for goal specification and problem decomposition, whether into sequential steps or hierarchical structures. We will cover methods to embed operational constraints and user preferences within prompts to shape an agent's planning. You will also examine strategies for iterative planning and re-planning, enabling agents to adapt to new information or unexpected issues. Additionally, the chapter details how to prompt agents to assess the viability of their plans and to oversee the broader task execution process. This knowledge is reinforced through practical application, where you will engineer prompts to direct an agent in creating a detailed, multi-step plan.