Back to Details

lovstudio-style-clone

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

style-clone — 文风克隆

style-clone — Writing Style Cloning

输入一篇或多篇样本文章，提取文风画像（结构化描述），然后将任意目标文章改写成该文风。
Input one or more sample articles to extract a writing style profile (structured description), then rewrite any target article in this exact style.

When to Use

When to Use

  • 用户提供样本文章，想分析/提取其文风
  • 用户想用某人（作家、博主、品牌）的文风改写自己的内容
  • 用户说"模仿 XX 风格"、"用这篇文章的文风改写"
  • 用户想统一团队/品牌的写作风格
  • Users provide sample articles and want to analyze/extract their writing style
  • Users want to rewrite their content in the writing style of a specific person (writer, blogger, brand)
  • Users say "Imitate XX's style", "Rewrite in the style of this article"
  • Users want to standardize the writing style of a team/brand

Workflow (MANDATORY)

Workflow (MANDATORY)

按顺序执行，不得跳步。
Execute in order, no skipping steps.

Step 1: 收集样本 & 目标

Step 1: Collect Samples & Target Article

用 
AskUserQuestion
确认：
问题：
1. 样本文章是否已在消息中提供？（若无，提示用户粘贴）
2. 要改写的目标文章是什么？（若无，只做文风分析输出画像）
3. 改写时是否保留原文结构，还是可以自由重组？
若用户在调用时已提供样本文章 + 目标文章，跳过提问直接执行。
Use 
AskUserQuestion
to confirm:
Questions:
1. Have sample articles been provided in the message? (If not, prompt the user to paste them)
2. What is the target article to rewrite? (If not provided, only perform style analysis and output the profile)
3. Should the original structure be retained during rewriting, or can it be freely restructured?
If the user has provided both sample articles and the target article when invoking, skip the questions and proceed directly.

Step 2: 提取文风画像

Step 2: Extract Writing Style Profile

对样本文章进行深度分析，输出结构化文风画像（Style Profile）。
画像必须覆盖以下八个维度：
Conduct in-depth analysis of the sample articles and output a structured Writing Style Profile.
The profile must cover the following eight dimensions:

1. 句式节奏（Sentence Rhythm）

1. Sentence Rhythm

  • 句子平均长度（短句 <15字 / 中句 15-30字 / 长句 >30字）
  • 短句与长句的交替模式
  • 是否常用省略、破折号、感叹号等停顿手段
  • Average sentence length (short sentences <15 words / medium sentences 15-30 words / long sentences >30 words)
  • Alternation pattern of short and long sentences
  • Whether pausing devices such as ellipses, em dashes, exclamation marks are commonly used

2. 段落结构（Paragraph Structure）

2. Paragraph Structure

  • 每段平均句数
  • 开头句惯用模式（设问？叙述？金句？）
  • 结尾句惯用模式（总结？留悬念？行动号召？）
  • 是否有"一句段"（单句成段）习惯
  • Average number of sentences per paragraph
  • Usual pattern of opening sentences (rhetorical question? narration? punchline?)
  • Usual pattern of closing sentences (summary? cliffhanger? call to action?)
  • Whether there is a habit of "single-sentence paragraphs"

3. 词汇偏好（Vocabulary Palette）

3. Vocabulary Palette

  • 用词风格（口语化 / 书面 / 学术 / 混搭）
  • 高频词汇类型（动词密度、形容词风格）
  • 是否使用专业术语、英文夹杂、数字量化
  • 回避的词汇类型
  • Word choice style (colloquial / formal / academic / mixed)
  • High-frequency word types (verb density, adjective style)
  • Whether technical terms, mixed English, numerical quantification are used
  • Avoided word types

4. 叙事视角（Narrative Voice）

4. Narrative Voice

  • 人称（第一人称"我" / 第二人称"你" / 第三人称旁观 / 混合）
  • 与读者的距离感（亲密 / 专业 / 权威 / 平等）
  • 是否直接呼叫读者
  • Point of view (first-person "I" / second-person "you" / third-person observer / mixed)
  • Sense of distance from readers (intimate / professional / authoritative / equal)
  • Whether readers are directly addressed

5. 情感温度（Emotional Register）

5. Emotional Register

  • 整体情绪基调（冷静 / 热情 / 克制 / 激进 / 幽默）
  • 是否有情绪弧线（开头平静 → 结尾激昂等）
  • 对读者情绪的调动方式
  • Overall emotional tone (calm / enthusiastic / restrained / radical / humorous)
  • Whether there is an emotional arc (e.g., calm opening → passionate closing)
  • Ways to mobilize readers' emotions

6. 论证逻辑（Argumentation Style）

6. Argumentation Style

  • 论点呈现方式（先结论后论据 / 先铺垫后揭示 / 问题导向）
  • 是否大量用案例、数据、类比
  • 转折词、递进词使用习惯
  • Argument presentation method (conclusion first then evidence / build-up then reveal / problem-oriented)
  • Whether cases, data, analogies are extensively used
  • Usage habits of transition words and progressive words

7. 修辞手法（Rhetorical Devices）

7. Rhetorical Devices

  • 常用修辞（排比、对仗、设问、比喻、反复）
  • 是否有标志性开头/结尾句式
  • Commonly used rhetorical devices (parallelism, antithesis, rhetorical question, metaphor, repetition)
  • Whether there are iconic opening/closing sentence patterns

8. 整体基因（Style DNA）

8. Style DNA

  • 一句话定义这种文风（例："克制理性的深夜独白体"）
  • 最像哪位已知作者/媒体（参照系）
  • 禁忌：这种文风绝对不会出现什么
输出格式：
markdown
undefined
  • Define this style in one sentence (e.g., "Restrained and rational late-night monologue style")
  • Which known author/media it most resembles (reference frame)
  • Taboos: What this style will never include
Output Format:
markdown
undefined

文风画像：[样本标题/作者]

Writing Style Profile: [Sample Title/Author]

文风基因：[一句话定义] 参照系：[最接近的已知风格]
Style DNA: [One-sentence definition] Reference Frame: [Closest known style]

句式节奏

Sentence Rhythm

[具体描述 + 典型例句摘录]
[Specific description + excerpt of typical example sentences]

段落结构

Paragraph Structure

[具体描述]
[Specific description]

词汇偏好

Vocabulary Palette

[具体描述 + 高频词示例]
[Specific description + examples of high-frequency words]

叙事视角

Narrative Voice

[具体描述]
[Specific description]

情感温度

Emotional Register

[具体描述]
[Specific description]

论证逻辑

Argumentation Style

[具体描述]
[Specific description]

修辞手法

Rhetorical Devices

[具体描述 + 典型例句]
[Specific description + typical example sentences]

禁忌清单

Taboo List

  • 不用 [具体词汇/句式]
  • 不做 [具体行为]

---
  • Do not use [specific words/sentence patterns]
  • Do not engage in [specific behaviors]

---

Step 3: 改写目标文章

Step 3: Rewrite Target Article

基于 Step 2 的文风画像，对目标文章进行改写。
改写原则：
  1. 保留语义：原文的核心信息、论点、事实必须完整保留
  2. 替换外壳：句式、词汇、节奏、段落结构、修辞全部换成样本文风
  3. 结构自由度：根据用户在 Step 1 的选择决定是否重组段落
  4. 禁忌遵守：严格遵守画像中的禁忌清单
改写后输出：
markdown
undefined
Rewrite the target article based on the writing style profile from Step 2.
Rewriting Principles:
  1. Preserve Semantics: The core information, arguments, and facts of the original text must be fully retained
  2. Replace Stylistic Shell: Sentence patterns, vocabulary, rhythm, paragraph structure, and rhetorical devices are all replaced with the sample style
  3. Structural Freedom: Decide whether to restructure paragraphs based on the user's choice in Step 1
  4. Abide by Taboos: Strictly follow the taboo list in the profile
Output after rewriting:
markdown
undefined

改写结果

Rewriting Result

[改写后全文]
[Full rewritten text]

关键文风调整说明（3-5条）

Key Style Adjustment Notes (3-5 points)

  • 原文 [XXX句式] → 改写为 [YYY句式]，因为样本文风偏好...
  • ...

---
  • Original [XXX sentence pattern] → rewritten to [YYY sentence pattern], because the sample style prefers...
  • ...

---

Step 4: 存档（可选）

Step 4: Save Profile (Optional)

若用户希望保存文风画像供后续复用，将画像写入：
./output/style-profiles/[作者或标签]-style-profile-[YYYY-MM-DD].md
If the user wishes to save the writing style profile for future reuse, write the profile to:
./output/style-profiles/[Author or Tag]-style-profile-[YYYY-MM-DD].md

多样本处理

Multi-Sample Processing

若用户提供多篇样本：
  1. 分别分析每篇，找出共同特征（在所有样本中都出现的）
  2. 记录个性特征（只在部分样本出现，标注为"偶发"）
  3. 最终画像以共同特征为核心，偶发特征列为"可选用"
If the user provides multiple samples:
  1. Analyze each sample separately and identify common features (present in all samples)
  2. Record individual features (present only in some samples, marked as "occasional")
  3. The final profile takes common features as the core, and occasional features are listed as "optional"

注意事项

Notes

  • 文风画像是描述性的，不是评判性的——不说"好"或"差"
  • 若样本文章版权敏感，不直接大段引用，只摘取 <20字 的典型例句用于说明
  • 改写长文（>3000字）时，优先改写前 1/3 作为样本给用户确认，再继续
  • The writing style profile is descriptive, not evaluative—do not use words like "good" or "bad"
  • If the sample article has copyright sensitivity, do not directly quote large sections; only extract typical example sentences of <20 words for illustration
  • When rewriting long articles (>3000 words), prioritize rewriting the first 1/3 as a sample for user confirmation before proceeding