What it does
支持小红书自动发布、自动评论、自动检索的 Skill。支持 OpenClaw、Codex、CC 等
- Connects Instruction files the AI reads, with steps that call an outside service
- Account or key Some steps need a key for XSEC_TOKEN (per the README)
- Category Other
For makers: add a badge to your README
If you made this, paste this code into your README. The badge links back to this page.
[](https://mcp.popnx.com/en/skills/white0dew-xiaohongshuskills)Similar Skills
Same kind of work, ordered like the Skills list
| This one XiaohongshuSkills | llmfit | claude-video | VideoCaptioner | claude-cookbooks | |
|---|---|---|---|---|---|
| Maker | white0dew | AlexsJones | bradautomates | WEIFENG2333 | anthropics |
| Skills | 1 | 1 | 1 | 1 | 4 |
| GitHub stars | 3,619 | 39,096 | 21,249 | 16,577 | 54,867 |
| Last code change | Nov 18, 2026 | Nov 18, 2026 | Sep 25, 2026 | Sep 12, 2026 | Nov 17, 2026 |
| License | MIT | MIT | MIT | GPL-3.0 | MIT |
1 skills (instruction sets) by white0dew that an AI reads and follows · not an MCP server
How a Skill differs from MCP
MCP = the hands
Connects the AI to an outside service or data, such as reading Notion or creating an invoice. A program or server has to be running.
Skill = the manual
A text file that teaches the AI how to do a task. It connects to nothing.
Some overlap: a few Skills run scripts that call tools, and a few MCP servers only send knowledge. Start from the task you want done.
Works with
Claude CodeOpenCode
From the maker’s README and the folders in the repo
Install
Is it safe?
- Who made it
GitHub account white0dew (person), 7 years old
- Use and upkeep
★ 3,619 stars · code last changed Nov 18, 2026
- Any scripts?
8 scripts that can run on a computer: scripts/account_manager.py, scripts/cdp_publish.py, scripts/chrome_launcher.py, scripts/feed_explorer.py and 4 more
- Uses the internet?
Scripts that go online: scripts/image_downloader.py, scripts/generate_star_history.py, scripts/chrome_launcher.py, scripts/cdp_publish.py
- Changes machine-wide settings?
No hooks or writes to machine settings found
Checked automatically from the repo files at commit 0c89ba7 (read only, nothing was run). Not a safety guarantee: read it and decide for yourself
First: which AI do you use?
Pick your app and this page shows only how to install it there
Option 1Upload to claude.ai (web and app)
Files are byte-for-byte the same as on GitHub at commit 0c89ba7 · View on GitHub ↗
8 scripts
Once uploaded it runs on Claude’s systems, not on your computer. No need to unzip
Look inside before downloading
All 20 files · 1.4 MB
AGENTS.md2 KBLICENSE1 KBREADME.md18 KBSKILL.md14 KBassets/apimart-banner.png993 KBassets/star-history.svg8 KBconfig/accounts.json.example1 KBdocs/claude-code-integration.md3 KBdocs/code-review-2026-03-07.md9 KBpublic/whitedew.jpg112 KBrequirements.txt1 KBscripts/account_manager.py9 KBscripts/cdp_publish.py188 KBscripts/chrome_launcher.py11 KBscripts/feed_explorer.py28 KBscripts/generate_star_history.py7 KBscripts/image_downloader.py7 KBscripts/publish_pipeline.py22 KBscripts/run_lock.py3 KBtodo.md1 KB
Download redbookskills.zip
Open Skills on claude.ai and upload
+Create skillUpload a skillpick redbookskills.zip
No Upload button? Turn on file creation first
SettingsCapabilitiesCode execution and file creation
Turn on Code execution and file creation first (newer menus: Create and edit files) · personal plans: Settings → Capabilities · Team/Enterprise: an admin turns it on under Organization settings → Plugins & skills · every plan, including Free
Upload once; it works in claude.ai and Claude Desktop, and in recent Claude Code signed in with a Claude account (not with an API key)
Cowork and cloud sessions use the skills turned on in your claude.ai account, not ~/.claude/skills on your computer (Claude Desktop scheduled tasks run locally and still load it)
Checked against official docs Nov 12, 2026 · Source: support.claude.com, support.claude.com, code.claude.com
Option 1Let Claude Code install it
Copy this sentence and paste it into a Claude Code chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Restart Claude Code, then type /XiaohongshuSkills or ask for a matching task and Claude picks it up
Checked against official docs Nov 12, 2026 · Source: code.claude.com
For Terminal users
A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.
Open Terminal
Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead
Copy the skill folder to ~/.claude/skills
This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md
git clone --depth 1 https://github.com/white0dew/xiaohongshuskills.git /tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7 && mkdir -p ~/.claude/skills && cp -R '/tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7/.' ~/.claude/skills/xiaohongshuskillsRestart Claude Code, then type /XiaohongshuSkills or ask for a matching task and Claude picks it up
ChatGPT desktop app
The ChatGPT desktop app has Codex built in, and Codex reads skills from ~/.agents/skills
Option 1Let ChatGPT install it
Open Codex in the ChatGPT desktop app and paste this sentence into the chat
The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Option 2Put the folder there yourself
Put the redbookskills folder at ~/.agents/skills/redbookskills
~ is your home folder ($HOME)
Without Terminal: download the ZIP, unzip it and move the redbookskills folder to ~/.agents/skills
For Terminal users
A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.
Open Terminal
Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead
Copy the skill folder to ~/.agents/skills
This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md
git clone --depth 1 https://github.com/white0dew/xiaohongshuskills.git /tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7 && mkdir -p ~/.agents/skills && cp -R '/tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7/.' ~/.agents/skills/xiaohongshuskills
After installing
- ChatGPT desktop app: open Skills in the sidebar to see that the skill is there
- To use it: in ChatGPT type @ and pick the skill · in Codex CLI or the IDE extension run /skills or type $ and the skill name
- Codex picks up newly installed skills by itself; if one does not show up, restart Codex
Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com
ChatGPT on the web and mobile
Skills you install yourself do not work here; only skills that come with a plugin from the directory
To use this skill with ChatGPT, use the ChatGPT desktop app as above
Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com
Option 1Let Codex install it
Copy this sentence and paste it into a Codex chat · Works in Codex inside the ChatGPT desktop app, Codex CLI and the IDE extension
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Codex picks up newly installed skills by itself; if one does not show up, restart Codex · To use it: in ChatGPT type @ and pick the skill · in Codex CLI or the IDE extension run /skills or type $ and the skill name
Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com
For Terminal users
A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.
Open Terminal
Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead
Copy the skill folder to ~/.agents/skills
This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md
git clone --depth 1 https://github.com/white0dew/xiaohongshuskills.git /tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7 && mkdir -p ~/.agents/skills && cp -R '/tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7/.' ~/.agents/skills/xiaohongshuskillsRestart Codex, then ask for a matching task; Codex lists the skill
Option 1Let Cursor install it
Copy this sentence and paste it into a Cursor chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
In Cursor’s Agent chat, type / and pick XiaohongshuSkills
Checked against official docs Nov 12, 2026 · Source: cursor.com
For Terminal users
A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.
Open Terminal
Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead
Copy the skill folder to ~/.cursor/skills
This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md
git clone --depth 1 https://github.com/white0dew/xiaohongshuskills.git /tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7 && mkdir -p ~/.cursor/skills && cp -R '/tmp/popnx-skill-white0dew-xiaohongshuskills-0c89ba7/.' ~/.cursor/skills/xiaohongshuskillsIn Cursor’s Agent chat, type / and pick XiaohongshuSkills
Option 1Let the AI install it
For other apps that read skills. The sentence gives the paths for Claude Code, Codex and Cursor so the AI picks the one for your app · Copy this sentence into the app’s chat
For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing
The AI tells you what it will do first and asks your permission at every step
Checked against official docs Nov 12, 2026 · Source: code.claude.com, learn.chatgpt.com, cursor.com
For Terminal users
Other apps that support Agent Skills (such as Gemini CLI or GitHub Copilot) use different folders. Check that app’s guide, then copy the folder with the SKILL.md there.
Skill folders checked against the official Claude Code, Codex and Cursor guides on Nov 11, 2026
✓ Commands checked Nov 18, 2026
View the original SKILL.md
The text below comes from the maker’s repo, not from PopNx MCP
Showing the first 8,000 characters · Open the full file on GitHub
---
name: RedBookSkills
description: |
将图文/视频内容自动发布到小红书(XHS),并支持登录检查、内容检索与互动操作。
适用场景:发布图文、发布视频、仅启动测试浏览器、获取登录二维码、首页推荐抓取、搜索笔记、评论互动、抓取内容数据。
metadata:
trigger: 发布内容到小红书
source: Angiin/Post-to-xhs
---
# Post-to-xhs
你是“小红书发布助手”。目标是在用户确认后,调用本 Skill 的脚本完成发布或互动操作。
## 风险提示(重要)
**使用本 Skill 进行小红书自动化,存在被平台风控、限流、封号或封禁账号的风险。**
默认提醒用户优先使用测试号、小流量运行,并对最终内容进行人工复核。使用者需自行评估并承担相关风险。
## 输入判断
优先按以下顺序判断:
1. 用户明确要求"测试浏览器 / 启动浏览器 / 检查登录 / 获取登录二维码 / 只打开不发布":进入测试浏览器流程。
2. 用户要求“首页推荐 / 搜索笔记 / 找内容 / 查看某篇笔记详情 / 查看内容数据表 / 给帖子评论 / 回复评论 / 点赞收藏互动 / 查看用户主页 / 查看评论和@通知”:进入内容检索与互动流程(`list-feeds` / `search-feeds` / `get-feed-detail` / `post-comment-to-feed` / `respond-comment` / `note-upvote` / `note-unvote` / `note-bookmark` / `note-unbookmark` / `profile-snapshot` / `notes-from-profile` / `get-notification-mentions` / `content-data`)。
3. 用户已提供 `标题 + 正文 + 视频(本地路径或 URL)`:直接进入视频发布流程。
4. 用户已提供 `标题 + 正文 + 图片(本地路径或 URL)`:直接进入图文发布流程。
5. 用户只提供网页 URL:先提取网页内容与图片/视频,再给出可发布草稿,等待用户确认。
6. 信息不全:先补齐缺失信息,不要直接发布。
## 必做约束
- 发布前必须让用户确认最终标题、正文和图片/视频。
- 图文发布时,没有图片不得发布(小红书发图文必须有图片)。
- 视频发布时,没有视频不得发布。图片和视频不可混合使用(二选一)。
- 默认使用无头模式;若检测到未登录,切换有窗口模式登录。
- 标题长度不超过 38(中文/中文标点按 2,英文数字按 1)。
- 用户要求"仅测试浏览器"时,不得触发发布命令。
- 如使用文件路径,优先使用绝对路径;若用户给的是相对路径,先转换为绝对路径再执行命令。
- 若发布页结构异常,优先检查 `scripts/cdp_publish.py` 里的 `SELECTORS`、多图上传等待、正文编辑器与发布按钮点击逻辑;这些是最容易被小红书网页改版影响的区域。
## 测试浏览器流程(不发布)
1. 启动 post-to-xhs 专用 Chrome(默认有窗口模式,便于人工观察)。
2. 如用户要求静默运行,再使用无头模式。
3. 可选:执行登录状态检查并回传结果。
4. 结束后如用户要求,关闭测试浏览器实例。
## 图文发布流程
1. 准备输入(标题、正文、图片 URL 或本地图片)。
2. 如需文件输入,先写入 `title.txt`、`content.txt`。
3. 执行发布命令(默认无头)。
4. 回传执行结果(成功/失败 + 关键信息)。
## 视频发布流程
1. 准备输入(标题、正文、视频文件路径或 URL)。
2. 如需文件输入,先写入 `title.txt`、`content.txt`。
3. 执行视频发布命令(默认无头)。视频上传后需等待处理完成。
4. 回传执行结果(成功/失败 + 关键信息)。
## 内容检索与互动流程(搜索/详情/评论/内容数据)
1. 先检查小红书主页登录状态(`XHS_HOME_URL`,非创作者中心)。
2. 若用户需要首页推荐流,执行 `list-feeds` 获取首页推荐笔记列表。
3. 若用户需要关键词搜索,执行 `search-feeds` 获取笔记列表(默认会先抓取搜索下拉推荐词,结果字段为 `recommended_keywords`;当前返回页面可提取结果,如只需前 N 条由调用方自行截断,暂无单独 `--limit` 控制搜索结果条数)。
4. 若用户需要详情,从搜索结果中取 `id` + `xsecToken` 再执行 `get-feed-detail`;如用户明确要更多评论,可加 `--load-all-comments` 等参数。
5. 若用户需要发表评论,执行 `post-comment-to-feed`(一级评论;必填 `feed_id` / `xsec_token` / `content`)。
6. 若用户需要回复某条评论,执行 `respond-comment`(可用 `comment_id` / `comment_author` / `comment_snippet` 定位目标评论)。
7. 若用户需要点赞/收藏互动,执行 `note-upvote` / `note-unvote` / `note-bookmark` / `note-unbookmark`。
8. 若用户需要用户主页信息,执行 `profile-snapshot` 或 `notes-from-profile`。
9. 若用户需要“评论和@通知”,执行 `get-notification-mentions` 抓取 `/notification` 页面对应的 `you/mentions` 接口返回。
10. 若用户需要“笔记基础信息表”,执行 `content-data` 获取曝光/观看/点赞等指标。
11. 回传结构化结果(数量、核心字段、链接)。
## 常用命令
### 参数顺序提醒(`cdp_publish.py` / `publish_pipeline.py`)
请严格按下面顺序写命令,避免 `unrecognized arguments`:
- 全局参数放在子命令前:`--host --port --headless --account --timing-jitter --reuse-existing-tab`
- 子命令参数放在子命令后:如 `search-feeds` 的 `--keyword --sort-by --note-type`
- 常见可选全局参数:`--host 10.0.0.12 --port 9222 --reuse-existing-tab --account NAME`
示例(正确):
```bash
python scripts/cdp_publish.py --reuse-existing-tab search-feeds --keyword "春招" --sort-by 最新 --note-type 图文
```
### 0) 启动 / 测试浏览器(不发布)
默认 CDP 地址为 `127.0.0.1:9222`;可按需叠加 `--host` / `--port` 指向远程 Chrome。
```bash
# 启动测试浏览器(有窗口,推荐)
python scripts/chrome_launcher.py
# 可选:无头启动
python scripts/chrome_launcher.py --headless
# 检查当前登录状态
python scripts/cdp_publish.py check-login
# 常见变体:优先复用已有标签页
python scripts/cdp_publish.py --reuse-existing-tab check-login
# 远程 CDP 检查登录
python scripts/cdp_publish.py --host 10.0.0.12 --port 9222 check-login
# 获取登录二维码(返回 Base64,可供远程前端展示扫码)
python scripts/cdp_publish.py get-login-qrcode
# 重启 / 关闭测试浏览器
python scripts/chrome_launcher.py --restart
python scripts/chrome_launcher.py --kill
```
### 0.5) 首次登录 / 重新登录
```bash
# 本地 Chrome 登录
python scripts/cdp_publish.py login
# 远程 CDP 登录(不会自动重启远程 Chrome)
python scripts/cdp_publish.py --host 10.0.0.12 --port 9222 login
```
### 1) 准备 title.txt / content.txt
若用户给的是标题和正文,可先写入临时文件再执行命令:
```bash
printf '%s\n' '这里是标题' > /abs/path/title.txt
printf '%s\n' '这里是正文' > /abs/path/content.txt
```
### 2) 无头发布 or 有头预览 —— 使用图片 URL 发布
```bash
# 默认推荐:无头自动发布
python scripts/publish_pipeline.py --headless \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--image-urls "https://example.com/1.jpg" "https://example.com/2.jpg"
# 仅预览:停留在发布页人工确认
python scripts/publish_pipeline.py \
--preview \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--image-urls "https://example.com/1.jpg" "https://example.com/2.jpg"
# 常见变体:远程 CDP / 复用已有标签页
python scripts/publish_pipeline.py --host 10.0.0.12 --port 9222 --reuse-existing-tab \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--image-urls "https://example.com/1.jpg"
```
说明:当 `--host` 不是 `127.0.0.1/localhost` 时,脚本会跳过本地 `chrome_launcher.py` 的自动启动/重启逻辑。
说明:`publish_pipeline.py` 默认自动点击发布;如需停留在发布页人工确认,请加 `--preview`。
### 3) 无头发布 or 有头预览 —— 使用本地图片发布
```bash
# 本地图片发布
python scripts/publish_pipeline.py --headless \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--images "/abs/path/pic1.jpg" "/abs/path/pic2.jpg"
# WSL/远程 CDP + Windows/UNC 路径:跳过本地文件预校验
python scripts/publish_pipeline.py --headless \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--images "\\\\wsl.localhost\\Ubuntu\\home\\user\\pic1.jpg" \
--skip-file-check
```
说明:当控制端在 WSL 运行,且传入 Windows/UNC 路径(如 `\\wsl.localhost\...`)时,可加 `--skip-file-check`,避免 Linux 侧 `os.path.isfile()` 误判不存在。
说明:脚本会自动识别 `C:\...`、`\\wsl.localhost\...` 等 Windows/UNC 路径,并在传给 `DOM.setFileInputFiles` 时保留原始路径形态。
说明:若需要强制保留原始路径,也可显式加 `--preserve-upload-paths`。
### 3.5) 视频发布(本地视频文件 / 视频 URL)
```bash
# 本地视频文件
python scripts/publish_pipeline.py --headless \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--video "/abs/path/my_video.mp4"
# 视频 URL
python scripts/publish_pipeline.py --headless \
--title-file /abs/path/title.txt \
--content-file /abs/path/content.txt \
--video-url "https://example.com/video.mp4"
```
### 4) 多账号发布 / 切换
```bash
python scripts/cdp_publish.py list-accounts
python scripts/cdp_publish.py add-account work --alias "工作号"
python scripts/cdp_publish.py --port 9223 --account work login
python scripts/publish_pipeline.py --port 9223 --account work --headless --title-file /abs/path/title.txt --content-file /abs/path/content.txt --image-urls "https://example.com/1.jpg"
```
### 5) 搜索内容 / 获取笔记详情
```bash
# 首页推荐笔记
python scripts/cdp_publish.py list-feeds
# 搜索笔记
python scripts/cdp_publish.py search-feeds --keyword "春招"
# 常见变体:带筛选 + 复用标签页
python scripts/cdp_publish.py --reuse-existing-tab search-feeds --keyword "春招" --sort-by 最新 --note-type 图文
# 获取笔记详情(feed_id 与 xsec_token 来自搜索结果)
python scripts/cdp_publish.py get-feed-detail \
--feed-id 67abc1234def567890123456 \
--xsec-token XSEC_TOKEN
# 可选:滚动加载更多一级评论,并尝试展开二级回复
python scripts/cdp_publish.py get-feed-detail \
--feed-id 67abc1234def567890123456 \
--xsec-token XSEC_TOKEN \
--load-all-comments \
--limit 20 \
--click-more-replies \
--reply-limit 10 \
--scroll-speed normal
```
说明:`list-feeds` 返回首页推荐 feed 列表。
说明:`search-feeds` 输出中包含 `recommended_keywords_count` 与 `recommended_keywords`,表示回车前搜索框下拉推荐词。
说明:`search-feeds` 返回当前页面可提取到的结果,不提供单独的 `--limit` 条数控制;若只需前 N 条,请在调用方截断返回列表。
说明:`get-feed-detail --load-all-comments` 会先滚动评论区,并可选点击“更多回复”后再提取详情,同时额外返回 `comment_loading`。
说明:`check-login` 与主页登录检查默认启用本地缓存(12h,仅缓存“已登录”),到期后自动重新网页校验。
### 6) 给笔记发表评论(一级评论)
```bash
# 直接传评论文本
python scripts/cdp_publish.py post-comment-to-feed \
--feed-id 67abc1234def567890123456 \
--xsec-token XSEC_TOKEN \
--content "写得很实用,感谢分享"
# 使用文件传评论(适合多行文本)
python scripts/cdp_publish.py post-comment-to-feed \
--feed-id 67abc1234def567890123456 \
--xsec-token XSEC_TOKEN \
--content-file "/abs/path/comment.txt"
```
### 7) 获取内容数据表(content_data)
```bash
# 获取笔记基础信息表(曝光/观看/封面点击率/点赞/评论/收藏/涨粉/分享/人均观看时长/弹幕)
python scripts/cdp_publish.py content-data
# 下划线别名
python scripts/cdp_publish.py content_data
# 可选:导出 CSV
python Discussion
Ask, share a tip or report that it does not work. Sorted by what people marked helpful.
No posts yet. Be the first to ask or share.
Posts here are written by users, not the PopNx team.
From GitHub
- GitHub stars
- 3,619
- Maker
- white0dew
- License
- MIT
- Last code change
- Nov 18, 2026
- SKILL.md files in the repo (including translated copies and templates)
- 1