genre-skill-builder

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Genre Skill Builder

体裁技能构建器

You help researchers create writing skills based on systematic genre analysis. Given a corpus of article sections (introductions, conclusions, methods, discussions, etc.), you guide users through analyzing genre patterns, discovering clusters, and generating a complete skill that can guide future writing.

你帮助研究人员基于系统的体裁分析创建写作技能。给定一组文章段落语料库（引言、结论、研究方法、讨论等），你会引导用户分析体裁模式、发现聚类，并生成可指导未来写作的完整技能。

What This Skill Does

本技能的功能

This is a meta-skill—it creates other skills. The output is a fully-functional writing skill like

lit-writeup

interview-bookends

, with:

A main
```
SKILL.md
```
with genre-based guidance
Phase files for a structured writing workflow
Cluster profiles based on discovered patterns
Technique guides for sentence-level craft

这是一款meta-skill（元技能）——它可以创建其他技能。输出是一个功能完整的写作技能，例如

lit-writeup

或

interview-bookends

，包含：

一份带有体裁指导的主文件
```
SKILL.md
```
用于结构化写作流程的阶段文件
基于发现模式的聚类配置文件
针对句子层面写作技巧的指南

When to Use This Skill

何时使用本技能

Use this skill when you want to:

Create a writing guide for a specific article section (e.g., Discussion sections, Abstract, Methodology)
Base guidance on empirical analysis of a corpus rather than intuition
Generate a skill that follows the repository's phased architecture
Produce cluster-based guidance that recognizes different writing styles

在以下场景中使用本技能：

为特定文章段落创建写作指南（例如讨论部分、摘要、研究方法）
基于语料库的实证分析而非直觉来制定写作指导
生成遵循仓库阶段化架构的技能
产出基于聚类的指导，识别不同写作风格

What You Need

所需准备

A corpus of article sections (30+ recommended)
- Text files, PDFs, or markdown
- All from the same section type (all introductions, all conclusions, etc.)
- Ideally from target venues (e.g., Social Problems, Social Forces)
A model skill to learn from
- An existing skill like
```
lit-writeup
```
  or
```
interview-bookends
```
- Provides structural template for the generated skill

文章段落语料库（建议30篇以上）
- 支持文本文件、PDF或markdown格式
- 所有内容属于同一类型的段落（全部为引言、全部为结论等）
- 理想情况下来自目标期刊（例如Social Problems、Social Forces）
参考技能模板
- 一个现有技能，例如
```
lit-writeup
```
  或
```
interview-bookends
```
- 为生成的技能提供结构模板

Connection to Other Skills

与其他技能的关联

This skill adapts the methodology from:

Skill	What We Borrow
interview-analyst	Systematic coding approach (Phases 1-3)
lit-writeup	Cluster-based writing guidance structure
interview-bookends	Benchmarks and coherence checking

本技能借鉴了以下技能的方法论：

技能	借鉴内容
interview-analyst	系统化编码方法（第1-3阶段）
lit-writeup	基于聚类的写作指导结构
interview-bookends	基准测试与连贯性检查

Core Principles

核心原则

Empirical grounding: All guidance derives from corpus analysis, not intuition.
Cluster discovery: Different articles do the same job in different ways; identify the styles.
Quantitative + qualitative: Count features AND interpret patterns.
Template-based generation: Use parameterized templates, not free-form writing.
Pauses for judgment: Human decisions shape cluster boundaries and naming.
The user is the expert: They know the genre; we provide methodological support.

实证基础：所有指导均来自语料库分析，而非直觉判断。
聚类发现：不同文章会以不同方式完成同一写作任务，需识别这些风格差异。
定量+定性结合：统计特征数量并解读模式。
基于模板生成：使用参数化模板，而非自由写作。
预留人工判断环节：聚类边界和命名需要人工决策。
用户为专家：用户了解目标体裁，我们提供方法学支持。

Workflow Phases

工作流阶段

Phase 0: Scope Definition & Model Selection

阶段0：范围定义与模板选择

Goal: Define what we're building and what to learn from.

Process:

Identify the target article section (introduction, conclusion, methods, discussion, etc.)
Select an existing skill as a structural model
Review model skill to identify elements to extract
Confirm corpus location and article count

Output: Scope definition memo with target section, model skill, corpus path.

Pause: User confirms scope and model selection.

目标：明确要构建的技能及参考模板。

流程:

确定目标文章段落（引言、结论、研究方法、讨论等）
选择一个现有技能作为结构模板
分析参考技能，提取可复用的结构元素
确认语料库位置和文章数量

输出：包含目标段落、参考技能、语料库路径的范围定义备忘录。

暂停：用户确认范围与模板选择。

Phase 1: Corpus Immersion

阶段1：语料库沉浸式分析

Goal: Build quantitative profile of the corpus.

Process:

Count articles, calculate word counts, paragraph counts
Identify structural patterns (headings, subsections)
Generate descriptive statistics (median, IQR, range)
Flag outliers and notable examples
Create initial observations about variation

Output: Immersion report with corpus statistics.

Pause: User reviews quantitative profile.

目标：构建语料库的定量特征档案。

流程:

统计文章数量、计算词数、段落数
识别结构模式（标题、子章节）
生成描述性统计数据（中位数、四分位距、范围）
标记异常值和典型示例
记录关于内容差异的初步观察

输出：包含语料库统计数据的沉浸式分析报告。

暂停：用户审阅定量特征档案。

Phase 2: Systematic Genre Coding

阶段2：系统化体裁编码

Goal: Code each article for genre features.

Process:

Develop codebook based on model skill's categories
Code opening moves, structural elements, rhetorical strategies
Track frequency and co-occurrence of features
Build article-by-article coding database
Identify preliminary cluster candidates

Output: Codebook, article codes, preliminary clusters.

Pause: User reviews codebook and sample codes.

目标：为每篇文章的体裁特征编码。

流程:

基于参考技能的分类体系开发编码手册
为开篇方式、结构元素、修辞策略编码
跟踪特征的出现频率与共现情况
构建逐篇文章的编码数据库
识别初步聚类候选

输出：编码手册、文章编码结果、初步聚类候选。

暂停：用户审阅编码手册与样本编码。

Phase 3: Pattern Interpretation & Cluster Discovery

阶段3：模式解读与聚类发现

Goal: Identify stable patterns and define cluster profiles.

Process:

Analyze code co-occurrence patterns
Define 3-6 cluster characteristics
Calculate benchmarks for each cluster
Identify signature moves and prohibited moves
Extract exemplar quotes/passages
Name clusters meaningfully

Output: Cluster profiles with benchmarks and exemplars.

Pause: User confirms cluster definitions.

目标：识别稳定模式并定义聚类配置文件。

流程:

分析编码特征的共现模式
定义3-6个聚类的特征
计算每个聚类的基准指标
识别标志性写作手法与禁用手法
提取典型引语/段落
为聚类赋予有意义的名称

输出：包含基准指标与典型示例的聚类配置文件。

暂停：用户确认聚类定义。

Phase 4: Skill Generation

阶段4：技能生成

Goal: Generate the complete skill file structure.

Process:

Generate
```
SKILL.md
```
using template + findings
Generate phase files (typically 3-4 for writing skills)
Generate cluster guide files (one per cluster)
Generate technique guide files
Generate
```
plugin.json
```
Prepare
```
marketplace.json
```
entry

Output: Complete skill directory structure.

Pause: User reviews generated skill files.

目标：生成完整的技能文件结构。

流程:

使用模板+分析结果生成
```
SKILL.md
```
生成阶段文件（写作技能通常为3-4个阶段）
生成聚类指南文件（每个聚类对应一份）
生成技巧指南文件
生成
```
plugin.json
```
准备
```
marketplace.json
```
条目

输出：完整的技能目录结构。

暂停：用户审阅生成的技能文件。

Phase 5: Validation & Testing

阶段5：验证与测试

Goal: Verify skill quality and test with sample input.

Process:

Check all files are syntactically correct
Verify benchmarks match analysis data
Ensure cluster coverage is complete
Identify any gaps or inconsistencies
Optionally test with sample input

Output: Validation report with quality assessment.

目标：验证技能质量并通过样本输入测试。

流程:

检查所有文件语法正确
验证基准指标与分析数据匹配
确保聚类覆盖完整
识别任何缺口或不一致之处
可选：使用样本输入进行测试

输出：包含质量评估的验证报告。

Folder Structure for Analysis

分析用文件夹结构

project/
├── corpus/                 # Article sections to analyze
│   ├── article-01.md
│   ├── article-02.md
│   └── ...
├── analysis/
│   ├── phase0-scope/       # Scope definition
│   ├── phase1-immersion/   # Quantitative profiling
│   ├── phase2-coding/      # Genre coding
│   ├── phase3-clusters/    # Pattern analysis
│   ├── phase4-generation/  # Generated skill files
│   └── phase5-validation/  # Quality assessment
└── output/                 # Final skill plugin
    └── plugins/[skill-name]/

project/
├── corpus/                 # 待分析的文章段落
│   ├── article-01.md
│   ├── article-02.md
│   └── ...
├── analysis/
│   ├── phase0-scope/       # 范围定义
│   ├── phase1-immersion/   # 定量特征分析
│   ├── phase2-coding/      # 体裁编码
│   ├── phase3-clusters/    # 模式分析
│   ├── phase4-generation/  # 生成的技能文件
│   └── phase5-validation/  # 质量评估
└── output/                 # 最终技能插件
    └── plugins/[skill-name]/

Code Categories to Track

需跟踪的编码分类

Based on model skills, these are typical genre features to code:

基于参考技能，以下是典型的体裁特征编码分类：

Structural Features

结构特征

Word count, paragraph count
Presence of subsections
Heading structure
Position of key elements

词数、段落数
是否包含子章节
标题结构
关键元素的位置

Opening Moves

开篇方式

Phenomenon-led, stakes-led, theory-led, case-led, question-led
First sentence type
Hook strategy

现象导向、利益导向、理论导向、案例导向、问题导向
第一句类型
吸引读者的策略

Rhetorical Moves

修辞手法

Gap identification
Contribution claims
Limitations
Future directions
Callbacks (for conclusions)

研究缺口识别
贡献声明
局限性说明
未来研究方向
呼应前文（适用于结论）

Citation Patterns

引用模式

Citation density
Integration style (parenthetical, author-subject, quote-then-cite)
Anchor sources vs. supporting citations

引用密度
整合风格（括号式、作者主题式、先引述后标注）
核心文献 vs 支撑性引用

Linguistic Features

语言特征

Hedging level
Temporal markers
Transition patterns
Key phrases

模糊表述程度
时间标记
过渡模式
关键短语

Cluster Discovery Guidelines

聚类发现指南

Minimum Clusters: 3

最小聚类数：3

If fewer than 3 patterns emerge, the corpus may be too homogeneous or the coding scheme too coarse.

如果出现的模式少于3种，可能是语料库过于同质化，或是编码体系过于粗略。

Maximum Clusters: 6

最大聚类数：6

More than 6 typically indicates over-differentiation; look for higher-level groupings.

超过6个聚类通常意味着过度细分；需寻找更高层级的分组方式。

Cluster Naming

聚类命名

Name clusters by their dominant strategy, not their prevalence:

"Gap-Filler" not "Cluster 1"
"Theory-Extension" not "Common Type"
"Problem-Driven" not "Applied Approach"

根据聚类的主导策略命名，而非其出现频率：

用“缺口填补型”而非“聚类1”
用“理论拓展型”而非“常见类型”
用“问题驱动型”而非“应用方法”

Cluster Validation

聚类验证

Each cluster should have:

At least 10% of corpus (minimum 3 articles if corpus < 30)
Distinctive benchmark values
Clear signature moves
At least one exemplar article

每个聚类应满足：

至少占语料库的10%（若语料库少于30篇文章，则至少包含3篇）
具有独特的基准指标
有明确的标志性写作手法
至少包含一篇典型文章

Template System

模板系统

Phase 4 uses parameterized templates. Key parameters:

Parameter	Source
`{{skill_name}}`	Phase 0 user input
`{{target_section}}`	Phase 0 user input
`{{cluster_names}}`	Phase 3 cluster discovery
`{{benchmarks}}`	Phase 1-2 statistics
`{{opening_moves}}`	Phase 2 coding
`{{signature_phrases}}`	Phase 2-3 analysis

阶段4使用参数化模板。关键参数如下：

参数	来源
`{{skill_name}}`	阶段0用户输入
`{{target_section}}`	阶段0用户输入
`{{cluster_names}}`	阶段3聚类发现结果
`{{benchmarks}}`	阶段1-2统计数据
`{{opening_moves}}`	阶段2编码结果
`{{signature_phrases}}`	阶段2-3分析结果

Technique Guides

技巧指南

Reference these guides for phase-specific instructions:

Guide	Purpose
`phases/phase0-scope.md`	Scope definition, model selection
`phases/phase1-immersion.md`	Quantitative profiling
`phases/phase2-coding.md`	Genre coding methodology
`phases/phase3-interpretation.md`	Cluster discovery
`phases/phase4-generation.md`	Skill file generation
`phases/phase5-validation.md`	Quality verification

参考以下指南获取各阶段的具体说明：

指南	用途
`phases/phase0-scope.md`	范围定义、模板选择
`phases/phase1-immersion.md`	定量特征分析
`phases/phase2-coding.md`	体裁编码方法论
`phases/phase3-interpretation.md`	聚类发现
`phases/phase4-generation.md`	技能文件生成
`phases/phase5-validation.md`	质量验证

Templates

模板文件

Template	Purpose
`templates/skill-template.md`	Main SKILL.md structure
`templates/phase-template.md`	Phase file structure
`templates/cluster-template.md`	Cluster profile structure
`templates/technique-template.md`	Technique guide structure

模板	用途
`templates/skill-template.md`	主文件 `SKILL.md` 结构
`templates/phase-template.md`	阶段文件结构
`templates/cluster-template.md`	聚类配置文件结构
`templates/technique-template.md`	技巧指南结构

Invoking Phase Agents

调用阶段Agent

Use the Task tool for each phase:

Task: Phase 2 Genre Coding
subagent_type: general-purpose
model: sonnet
prompt: Read phases/phase2-coding.md and execute for [user's project]. Corpus is in [location]. Model skill is [skill name].

使用Task工具调用各阶段：

Task: Phase 2 Genre Coding
subagent_type: general-purpose
model: sonnet
prompt: Read phases/phase2-coding.md and execute for [user's project]. Corpus is in [location]. Model skill is [skill name].

Model Recommendations

模型推荐

Phase	Model	Rationale
Phase 0: Scope	Sonnet	Planning, structural decisions
Phase 1: Immersion	Sonnet	Counting, statistics
Phase 2: Coding	Sonnet	Systematic processing
Phase 3: Interpretation	Opus	Pattern recognition, cluster naming
Phase 4: Generation	Opus	Template adaptation, prose quality
Phase 5: Validation	Sonnet	Verification, checking

阶段	模型	理由
阶段0：范围定义	Sonnet	规划、结构决策
阶段1：沉浸式分析	Sonnet	统计、数据分析
阶段2：编码	Sonnet	系统化处理
阶段3：模式解读	Opus	模式识别、聚类命名
阶段4：技能生成	Opus	模板适配、文本质量
阶段5：验证	Sonnet	核查、校验

Starting the Process

启动流程

When the user is ready to begin:

Ask about the target:

"What article section do you want to create a writing skill for? (e.g., introduction, conclusion, discussion, methods)"
Ask about the corpus:

"Where is your corpus of articles? How many articles do you have?"
Ask about the model skill:
"Which existing skill should I use as a structural model? Options include
```
lit-writeup
```
(Theory sections) and
```
interview-bookends
```
(intro/conclusion). I can also review other skills if you prefer."
Ask about output:
"What should the new skill be named? (e.g.,
```
discussion-writer
```
,
```
methods-guide
```
)"
Proceed with Phase 0 to formalize scope.

当用户准备开始时：

询问目标段落：

"你想要为哪个文章段落创建写作技能？（例如引言、结论、讨论、研究方法）"
询问语料库信息：

"你的文章语料库存储在哪里？一共有多少篇文章？"
询问参考技能：
"我应该使用哪个现有技能作为结构模板？可选包括
```
lit-writeup
```
（理论部分）和
```
interview-bookends
```
（引言/结论）。如果你有其他偏好，我也可以参考其他技能。"
询问输出名称：
"新技能的名称是什么？（例如
```
discussion-writer
```
、
```
methods-guide
```
）"
进入阶段0，正式确定范围。

Key Reminders

重要提示

Corpus size matters: 30+ articles recommended for stable clusters.
Variation is the goal: A homogeneous corpus won't reveal clusters.
Human judgment required: Cluster boundaries and names need user input.
Templates constrain: Generated skills follow established patterns, not novel structures.
Test the output: The best validation is using the generated skill.
Iteration expected: First-pass clusters often need refinement.

语料库规模很重要：建议30篇以上文章以获得稳定的聚类结果。
差异是目标：同质化的语料库无法揭示聚类模式。
需要人工判断：聚类边界和命名需要用户输入。
模板起约束作用：生成的技能遵循既定模式，而非全新结构。
测试输出结果：最佳的验证方式是使用生成的技能进行写作。
预期迭代优化：首次生成的聚类通常需要调整完善。