Claude Skills
CollectionsCompareWorkflowsNominate
Sign inSign up
© 2026 Curated Agent Skills·Learn more about Agent Skills
Back to repository

beauty-step1

verified

Document content analysis and merging. Automatically invoked during step 1 of the beauty command to fully understand source document content, extract key information, and establish content structure. 文档内容分析合并。在beauty命令的步骤1执行时自动调用,用于完整理解源文档内容,提取关键信息,建立内容结构。

View on GitHub

Marketplace

within7-plugins-tools

Within-7/minto-plugin-tools

Plugin

html-presentation-beautifier

design

Repository
Verified Org

Within-7/minto-plugin-tools
1stars

html-presentation-beautifier/skills/beauty-step1/SKILL.md

Last Verified

February 2, 2026

Install Skill

Select agents to install to:

Scope:
npx add-skill https://github.com/Within-7/minto-plugin-tools/blob/main/html-presentation-beautifier/skills/beauty-step1/SKILL.md -a claude-code --skill beauty-step1

Installation paths:

Claude
.claude/skills/beauty-step1/
Powered by add-skill CLI

Instructions

# Beauty 步骤1:文档内容分析合并 / Beauty Step 1: Document Content Analysis

## 目标 / Goal

完整理解源文档内容,提取关键信息,建立内容结构。

Fully understand source document content, extract key information, and establish content structure.

## ⚠️ Token限制处理机制 / Token Limit Handling Mechanism

**如果文档过长无法一次性读取完整 / If document too long to read at once:**

```
正确做法(使用继续机制)/ Correct Approach (Use Continue Mechanism):

步骤1.1:读取文档前半部分 / Step 1.1: Read first half of document
├─ 使用 Read 工具读取文档(offset: 0, limit: 500)
   Use Read tool to read document (offset: 0, limit: 500)
├─ 提取前半部分的章节结构
   Extract chapter structure from first half
├─ 记录前半部分的数据点
   Record data points from first half
└─ 输出:"步骤1.1完成 - 已分析文档前半部分(0-500行)
       请输入'继续'以分析后半部分"
   Output: "Step 1.1 complete - Analyzed first half of document (lines 0-500)
           Please input 'continue' to analyze the second half"

【等待用户输入"继续" / Wait for user to input "continue"】

步骤1.2:读取文档后半部分 / Step 1.2: Read second half of document
├─ 使用 Read 工具读取文档(offset: 500, limit: 500)
   Use Read tool to read document (offset: 500, limit: 500)
├─ 提取后半部分的章节结构
   Extract chapter structure from second half
├─ 记录后半部分的数据点
   Record data points from second half
└─ 输出:"步骤1.2完成 - 已分析文档后半部分(500-1000行)
       文档分析100%完成"
   Output: "Step 1.2 complete - Analyzed second half of document (lines 500-1000)
           Document analysis 100% complete"

【如果文档超过1000行,继续分片读取 / If document exceeds 1000 lines, continue segmented reading】

步骤1.3:合并分析结果 / Step 1.3: Merge analysis results
├─ 整合所有部分的分析结果
   Integrate analysis results from all parts
├─ 生成完整的章节结构清单
   Generate complete chapter structure list
├─ 生成完整的数据点清单
   Generate complete data point list
└─ 进入步骤2
   Proceed to Step 2
```

**❌ 禁止做法(偷工减料)/ Prohibited Actions (Cutting Corners):**
```
错误做法1:只读前几行 / Wrong Approach 1: Only read first few lines
├─ Read 文件(limit: 50)
   Read file (limit: 50)
└─ ❌ 跳过了后面的内容
   ❌ Skipped remaining content

错误做法2:使用摘要 / Wrong Approach 2: Use summaries
├─ 请用户总结文档内容
   Ask user to summarize document conten

Validation Details

Front Matter
Required Fields
Valid Name Format
Valid Description
Has Sections
Allowed Tools
Instruction Length:
3966 chars