← Back to Skills Marketplace
wscats

小红书

by enoyao · GitHub ↗ · v1.0.10 · MIT-0
cross-platform ⚠ suspicious
7673
Downloads
15
Stars
0
Active Installs
10
Versions
Install in OpenClaw
/install xhs-cn
Description
小红书自动化运营工具-支持:搜索笔记、查看笔记详情及评论、浏览推荐流、发布图文笔记。 当用户提及 xiaohongshu、小红书、RedNote,或需要在该平台进行内容调研/发布时使用。
Usage Guidance
Install only if you are comfortable auditing the local MCP component yourself, using a dedicated Xiaohongshu account, and keeping the endpoint strictly local. Do not set XHS_ALLOW_REMOTE=yes or XHS_MCP_BASE_URL to a non-loopback address unless you fully trust that server and understand that tokens and draft content may be sent there. Review scripts/publish.sh only as a maintainer tool; normal users should not run it.
Capability Assessment
⚠ Credentials
The docs repeatedly describe loopback-only enforcement, but scripts/xhs_client.py allows non-loopback endpoints when XHS_ALLOW_REMOTE=yes is set. That can transmit search terms, xsec_token values, and unpublished post content to arbitrary HTTP(S) endpoints without an allowlist or HTTPS-only restriction.
ℹ Install Mechanism
Setup is manual and disclosed, including third-party component installation and a single Python dependency. The dependency is unpinned, which is a supply-chain hardening gap but not evidence of malicious installation behavior.
ℹ Instruction Scope
The activation text is broad, but the publishing workflow requires per-post preview and explicit authorization, and the client adds a dedicated-account gate plus interactive or --yes confirmation.
ℹ Persistence & Privilege
The skill itself does not install persistence. It relies on a third-party login/session component and tells users to stop the process and revoke the login device afterward. The included ClawHub publish script mutates registry state only if a maintainer explicitly sets CLAWHUB_PUBLISH_CONFIRM=yes.
ℹ Purpose & Capability
The main client matches the stated purpose: searching, reading, and publishing Xiaohongshu/RedNote content through a user-managed MCP component. Public posting is high impact but disclosed and guarded by preview and confirmation requirements.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install xhs-cn
  3. After installation, invoke the skill by name or use /xhs-cn
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.10
Version 1.0.10 - 增强客户端安全措施:新增自动检测,默认拒绝所有不安全或未授权的 `publish` 操作。 - SKILL.md 文档重写,明确所有客户端内置护栏,以及上游组件独立、非本仓库维护的提醒。 - 快速参考与市场调研、内容发布流程重构,突出强制逐步授权和机械性安全检查细节。 - 简化注意事项,删除/合并部分重复的人工安全警告,统一以客户端强制规约为准。
v1.0.7
Version 1.0.7 - 增加了“安装前知情同意(Informed Consent)”区块,首次使用前强调信任模型和安全边界。 - 明确告知用户自动化发布的敏感性,并建议仅用专用账号、本地部署且逐次授权操作。 - 现将 MCP 服务器的第三方来源、安全要求及会话托管风险写入顶部说明。 - 其余使用方法、指令与安全指引未更动。
v1.0.6
**重大更新:完善发布安全机制,防止误发,强化账号与隐私保护** - 发布流程新增“发布预览+逐次书面授权”环节,禁止模糊或静默同意,防止自动化脚本误发内容。 - 强调 `publish` 命令具公开不可撤回风险,建议使用专用账号,严禁主账号托管。 - 明确本地服务器仅允许 localhost 监听,安全性提升,杜绝误设暴露。 - 补充敏感操作须逐次获授权的硬性原则,禁止在无人值守环境调用发帖功能。 - 修订发布相关文档,区分 Skill 发布与小红书内容发布,注记脚本仅供开发者使用。 (此次更新无接口参数变更,兼容原有命令用法)
v1.0.5
- All documentation rewritten from English to Chinese for improved localization. - Usage instructions and workflows updated to use Chinese terminology, examples, and step-by-step guides. - Clarified command references and best practices in a fully localized format. - Updated descriptions to better reflect the skill’s use case and activation cues for Chinese-speaking users.
v1.0.4
- Added scripts/publish.sh to the repository. - No other functional or documentation changes.
v1.0.3
Version 1.0.3 of the xhs-cn skill is a major simplification and refactor: - Migrated to a lightweight wrapper model: skill now operates via a bundled Python client (scripts/xhs_client.py) that communicates with a required, separately-run local MCP server for Xiaohongshu automation. - Entire in-package Python SDK, examples, and most config files were removed; legacy code for direct automation is no longer included. - Documentation now focuses on command line workflows using the provided client script, with quick references for searching, publishing, and retrieval tasks. - Clear separation: all automation now handled through the external MCP server; setup instructions moved to SETUP.md.
v1.0.2
**Version 1.0.2 Changelog** - Added detailed skill manifest fields, including: `version`, `source`, `homepage`, `env_vars`, `local_files`, and `permissions` in SKILL.md. - Now lists and describes all optional environment variables for authentication and proxy/browser configuration. - Documents the handling and sensitivity of local files, especially `cookies.json` and `.env`. - Specifies permissions and operational scope of the skill. - README/guide content remains comprehensive and unchanged in feature scope.
v1.0.1
xhs-cn 1.0.1 introduces Python自动化操作与Playwright全链路真实小红书运营自动化。 - 新增 Playwright+异步API,支持二维码扫码自动登录、内容创作自动发布、互动评论等真实操作 - 新增 `examples/login_and_publish.py`、核心代码目录 `xiaohongshu/`,实现扫码登录/图文/视频发布、笔记搜索、点赞、评论等功能 - 配置/结构文档全面更新,详细说明业务能力、API、自动化流程及快速上手 - 支持 Cookie 持久化与自动复用,极大提升自动化体验 - 代码与文档均已模块化,支持二次开发与集成
v1.0.9
- 简化并精炼了技能描述文本,去除多余修饰,使介绍内容更加简洁直接。 - 删除了部分重复和啰嗦的说明,无更动功能和使用方式。 - 保持所有示例场景和功能模块不变,整体内容结构一致。 - 技能核心能力与原有版本保持一致,仅描述方式更加精炼。
v1.0.8
xiaohongshu 1.0.8 更新日志 - 新增SKILL.md,系统化介绍技能的功能、用法与应用场景。 - 明确罗列全链路运营能力,从内容创作、笔记发布、搜索与分析、评论互动到运营数据分析。 - 丰富使用示例,覆盖多风格生成、发布时间、竞品分析、引导互动等多种实际需求。 - 补充进阶用法与组合能力说明,提升实际使用效率。 - 增加注意事项,提醒用户内容合规与数据分析范围。
Metadata
Slug xhs-cn
Version 1.0.10
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 10
Frequently Asked Questions

What is 小红书?

小红书自动化运营工具-支持:搜索笔记、查看笔记详情及评论、浏览推荐流、发布图文笔记。 当用户提及 xiaohongshu、小红书、RedNote,或需要在该平台进行内容调研/发布时使用。 It is an AI Agent Skill for Claude Code / OpenClaw, with 7673 downloads so far.

How do I install 小红书?

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

Is 小红书 free?

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

Which platforms does 小红书 support?

小红书 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 小红书?

It is built and maintained by enoyao (@wscats); the current version is v1.0.10.

💬 Comments