Skip to content
TakoAPI
Back to skills

xiaohongshu-extract

by jovijovi

Coding Agents & IDEs
1.3kDownloads reported by clawskills.sh
1Stars
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Extract metadata from Xiaohongshu (XHS) share or discovery URLs by parsing window.__INITIAL_STATE__ and returning.

Description

What This Skill Does

Extracts note metadata from Xiaohongshu (XHS) share or discovery URLs by parsing the page's embedded state object. Returns structured JSON with title, description, engagement stats, user info, tags, and video stream details.

Parses the embedded window.__INITIAL_STATE__ directly, avoiding the need for a formal API key or authenticated session.

When to Use It

  • Pulling engagement stats from a viral XHS post
  • Archiving note metadata before a post is deleted
  • Feeding XHS content data into a downstream analysis pipeline
  • Extracting video stream URLs from XHS discovery links
  • Batch collecting note details from a list of XHS share URLs

Example Workflow

INPUT

User asks: extract metadata from this XHS share link: https://www.xiaohongshu.com/explore/abc123

AGENT
  1. 1

    Run scripts/xiaohongshu_extract.py with the provided URL and --pretty flag

  2. 2

    Check if window.__INITIAL_STATE__ was found in the page response

  3. 3

    If not found, ask the user for a direct discovery URL instead

  4. 4

    Parse the returned JSON for title, engagement counts, tags, and video info

  5. 5

    Return the structured note metadata to the user

OUTPUT

JSON object with note_id, title, desc, interact counts, tags, and video stream details

Installation

clawhub install jovijovi/xiaohongshu-extract

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links