xhs-batch

Original：🇨🇳 Chinese

Translated

Batch extract Xiaohongshu posts and organize them into a local Markdown knowledge base

1installs

Sourcechenxiachan/xhs-claude-skills

Added on2026-06-12

NPX Install

npx skill4agent add chenxiachan/xhs-claude-skills xhs-batch

SKILL.md Content (Chinese)

View Translation Comparison →

Users want to batch extract multiple Xiaohongshu posts. Please follow the steps below to process:

Constant Definitions

Cookies File:
```
~/cookies.json
```
Obsidian Save Directory:
```
~/Documents/Obsidian Vault/xhs
```
Whisper Model:
```
mlx-community/whisper-large-v3-turbo
```

Input

Link list provided by users: $ARGUMENTS

Workflow

Step 1: Parse Links

Extract all Xiaohongshu links from the input (supports multi-line, space-separated, comma-separated formats). Extract post ID and xsec_token from each link.

Step 2: Check Cookies

Check if

~/cookies.json

exists. If not, guide users to export it according to Step 0 of

/xhs

.

Step 3: Extract One by One

For each link, execute the complete extraction workflow of

/xhs

(Steps 2-4):

Request page → Parse INITIAL_STATE
Transcribe audio for video posts
Organize in Peter Thiel's style
Save as
```
{YYYY-MM-DD} {Short Title}.md
```

Wait 3 seconds between each post to avoid triggering anti-crawling mechanisms.

Step 4: Generate Summary Report

After all tasks are completed, output a brief summary:

Number of successful/failed extractions
File name and one-sentence summary for each post