xhs-analyze

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

用户希望对已提取的小红书收藏做 AI 分析。

The user wants to perform AI analysis on the extracted Xiaohongshu collected content.

常量定义

Constant Definitions

Obsidian 保存目录:
```
~/Documents/Obsidian Vault/xhs
```

Obsidian Save Directory:
```
~/Documents/Obsidian Vault/xhs
```

输入

Input

用户查询: $ARGUMENTS

User Query: $ARGUMENTS

流程

Process

步骤 1：加载数据

Step 1: Load Data

读取

<Obsidian 保存目录>

下所有

.md

文件（排除 img/video 子目录）。

如果用户指定了关键词，在文件名和内容中搜索匹配的帖子。如果参数为空，读取全部帖子进行总览分析。

Read all

.md

files under

<Obsidian Save Directory>

(exclude img/video subdirectories).

If the user specifies keywords, search for matching posts in file names and content. If the parameters are empty, read all posts for overview analysis.

步骤 2：AI 分析

Step 2: AI Analysis

根据内容类型和用户意图，选择合适的分析方式：

如果是教程/攻略类内容：

提炼核心步骤为编号清单
如有多篇同类内容，做对比总结

如果是知识/科普类内容：

提炼关键知识点
整理为结构化笔记

总览分析（无参数时）：

收藏内容的主题分布
高频标签和关注领域
推荐可以深入整理的主题

Select an appropriate analysis method based on content type and user intent:

If it is tutorial/guide content:

Extract core steps into a numbered list
If there are multiple similar contents, conduct a comparative summary

If it is knowledge/popular science content:

Extract key knowledge points
Organize into structured notes

Overview Analysis (when no parameters are provided):

Theme distribution of collected content
High-frequency tags and focus areas
Recommend themes that can be further organized

步骤 3：输出格式

Step 3: Output Format

用清晰的中文 Markdown 格式输出分析结果。

Output the analysis results in clear Chinese Markdown format.