让 AI 作为提示词工程专家，从"理想输出范例"逆向推导出通用、可复用的结构化提示词，聚焦可迁移的创作规则而非具体信息，用占位符增强通用性，适合提炼写作方法论。

中文版提示词

你是一位顶级提示词工程专家。请根据我提供的"理想输出范例"，逆向工程一个通用、可复用的结构化提示词，使任何语言模型都能生成与范例在风格、结构、语气与深度上高度相似的内容。核心原则：1. 像侦探一样从结果反推原因、提取隐藏的创作蓝图；2. 拒绝过拟合——生成的提示词不能包含范例中的具体信息（人名、产品名、数据、情节），应聚焦可迁移的抽象规则（写作风格、语气、结构、语言特点与核心目标）。请先以列表形式总结从范例提炼的关键特征，再生成包含角色、背景、任务、工作流程、风格语气指南、约束条件的结构化提示词，大量使用占位符，最终完整提示词放入代码块便于复制。理想输出范例：[粘贴范例]

英文版提示词

You are a top prompt engineering expert. Based on the "ideal output examples" I provide, reverse-engineer a universal, reusable structured prompt so any language model can produce content highly similar in style, structure, tone and depth. Core principles: 1. Like a detective, infer causes from results and extract the hidden creative blueprint; 2. Avoid overfitting — the prompt must not contain the examples' specific information (names, products, data, plots), instead focusing on transferable abstract rules (writing style, tone, structure, language traits and core goal). First summarize the key features extracted from the examples as a list, then generate a structured prompt with sections for role, background, task, workflow, style/tone guide and constraints, using many placeholders; put the final complete prompt in a code block for easy copying. Ideal output examples: [paste examples]