← Back to Skills Marketplace
ken0521

Xiaohongshu Skill

by 多动朕 · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ suspicious
287
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install xiaohongshu-win
Description
小红书内容工具 Windows 原生版。 基于 Node.js + Playwright,直接控制本地 Chromium 浏览器, 无需 WSL、无需 Linux 二进制、无需 Python、无需任何外部服务。 核心功能: - 🔍 内容搜索 - 关键词搜索,分析热度排行 - 📊 话题报告 - 自动生成热点分析...
Usage Guidance
This skill appears to do what it says: control a local Chromium via Playwright to search, read, report on, and (optionally) publish Xiaohongshu content. Before installing, consider: (1) the tool will store login cookies and a persistent browser profile under %USERPROFILE%\.xiaohongshu-win — those files contain your session and should be protected; (2) publishing actions are supported — review and test publish flows in a safe account to avoid accidental posts; (3) Playwright will download a ~150MB Chromium binary from the Playwright download mirror unless you point it elsewhere; (4) SKILL.md mentions an AI cover-generation integration not visible in the shown code — if that feature is important, ask the author where that integration runs and whether it contacts an external API; (5) run the script from an account/workspace you trust (the author warns about paths with non-ASCII characters). If you need stronger assurance, review the remaining/truncated portions of xhs-core.js (publish/checkStatus implementations) and verify there are no unexpected network endpoints or telemetry before granting continued use.
Capability Analysis
Type: OpenClaw Skill Name: xiaohongshu-win Version: 1.1.0 The skill is a Windows-native automation tool for Xiaohongshu (XHS) using Node.js and Playwright. It provides functionality for searching content, generating reports, and publishing notes by automating a Chromium browser instance. The code manages login sessions via local cookie storage in the user's profile directory (`.xiaohongshu-win`) and includes a migration script (`migrate-and-test.js`) to transition data from older versions. While it uses techniques to bypass anti-bot detection (e.g., disabling `AutomationControlled` and injecting cookies into the browser context), these actions are consistent with the stated purpose of social media automation and do not exhibit signs of malicious intent or data exfiltration.
Capability Assessment
Purpose & Capability
Name/description (Windows Xiaohongshu tool using Node.js + Playwright) match the included scripts (xhs-core.js and xhs.js). The code implements login, search, detail, publish, and track features described in SKILL.md. No unrelated cloud credentials, binaries, or external services are required by the code.
Instruction Scope
Runtime instructions and SKILL.md are consistent with the code: install playwright, download Chromium, run commands from a non-Chinese path, and use login which opens a browser for manual authentication. The skill reads/writes to %USERPROFILE%\.xiaohongshu-win (cookies, browser profile, reports) which is appropriate for storing session/profile data. One minor mismatch: SKILL.md mentions an '即梦AI' cover-generation integration, but no obvious references to an external AI service appear in the visible code (the core file was truncated in the listing—this could be implemented elsewhere).
Install Mechanism
There is no special install spec bundled; the project relies on npm (playwright) and npx playwright install chromium as documented—standard for Playwright. Playwright will download Chromium from its configured host (SKILL.md suggests using npm mirror if needed). No custom download URLs, shorteners, or unexpected installers are present in the repo files provided.
Credentials
The skill declares no required env vars but the code uses common environment variables: %USERPROFILE% for data storage and optional PLAYWRIGHT_BROWSERS_PATH / PLAYWRIGHT_DOWNLOAD_HOST for browser location/download mirror. It stores and updates site cookies (cookies.json) and a browser profile in the user's home directory—this is expected behavior for a browser automation tool, but users should understand these files contain session tokens and persistent profile data.
Persistence & Privilege
The skill does not request 'always: true' or elevated platform privileges. It persists data only under the user's profile directory (.xiaohongshu-win). Included migrate-and-test script will copy cookies from an older path and trigger a status check, which is consistent with a migration helper rather than an escalation of privileges.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install xiaohongshu-win
  3. After installation, invoke the skill by name or use /xiaohongshu-win
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
Version 1.1.0 highlights several new features and enhancements: - 新增定时任务功能,支持 Cron 定时自动搜热点内容 - 集成即梦AI,支持生成封面图片 - 支持 Browser Relay,可复用本地 Chrome 浏览器 - 新增 metadata 字段(author、tags) - 更新说明文档,强调核心功能与使用场景
v1.0.0
Initial release. Windows-native Xiaohongshu (RedNote) tool based on Node.js + Playwright. No WSL or Linux required. Supports search, post detail, publish notes, and topic reports on Windows 10/11.
Metadata
Slug xiaohongshu-win
Version 1.1.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Xiaohongshu Skill?

小红书内容工具 Windows 原生版。 基于 Node.js + Playwright,直接控制本地 Chromium 浏览器, 无需 WSL、无需 Linux 二进制、无需 Python、无需任何外部服务。 核心功能: - 🔍 内容搜索 - 关键词搜索,分析热度排行 - 📊 话题报告 - 自动生成热点分析... It is an AI Agent Skill for Claude Code / OpenClaw, with 287 downloads so far.

How do I install Xiaohongshu Skill?

Run "/install xiaohongshu-win" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Xiaohongshu Skill free?

Yes, Xiaohongshu Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Xiaohongshu Skill support?

Xiaohongshu Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Xiaohongshu Skill?

It is built and maintained by 多动朕 (@ken0521); the current version is v1.1.0.

💬 Comments