Prototype · Preview: every server in the official registry, synced every 30 minutes. Sign in with Google to write reviews.Prototype · updated every 30 min
PopNx MCPMCP servers for people who use AI · explained in plain words

xhs-writer-skill Skill

Repository ↗

by JuneYaooo★ 2121 skills

Install

What it does

小红书爆款笔记生成器,基于gpt-image-2,强力超快生成,自然语言修改,超好用!

  • Connects Instruction files the AI reads
  • Account or key Not confirmed yet (README not read yet)
  • Category Coding
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.

[![Featured on PopNx MCP](https://mcp.popnx.com/badge/skill/juneyaooo-xhs-writer-skill.svg)](https://mcp.popnx.com/en/skills/juneyaooo-xhs-writer-skill)

Use together with

MCP servers that connect the AI to services for the same work

REA

Reverse engineer anything from your terminal or agent with one CLI and MCP server.

by morlutoWidely used↓ 388.7K/wk★ 468KUpdated 7 days ago

MCP · Connects a serviceRuns on your computer

Fallow

Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.

by fallow-rsWidely used↓ 2.4M/wk★ 5.4KUpdated 2 days ago

MCP · Connects a serviceRuns on your computer

Desktop Commander

MCP server for terminal commands, file operations, and process management

by wonderwhy-erWidely used↓ 740.3K/wk★ 10.3KUpdated yesterday

MCP · Connects a serviceRuns on your computer

Similar Skills

Same kind of work, ordered like the Skills list

Compared with similar ones
This one
xhs-writer-skill
reaandrej-karpathy-skillsgraphifyMoneyPrinterTurbo
MakerJuneYaooomorlutomultica-aiGraphify-Labsharry0703
Skills11111
GitHub stars212467,235234,125149,873144,332
Last code changeApr 25, 2026Nov 18, 2026Apr 20, 2026Nov 18, 2026Nov 18, 2026
LicenseApache-2.0MIT–Apache-2.0MIT

1 skills (instruction sets) by JuneYaooo 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 CodeCodexCursor

From the maker’s README and the folders in the repo

Install

Is it safe?

  • Who made it

    GitHub account JuneYaooo (person), 6 years old

  • Use and upkeep

    ★ 212 stars, few users so far · code last changed Apr 25, 2026

  • Any scripts?

    8 scripts that can run on a computer: install_as_skill.sh, scripts/analyze_material.py, scripts/collage_3x4.py, scripts/crop_watermark.py and 4 more

  • Uses the internet?

    Scripts that go online: scripts/image_generator.py

  • Changes machine-wide settings?

    Scripts that touch settings folders (~/.claude or settings.json): install_as_skill.sh

Checked automatically from the repo files at commit 6adef3b (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)

Uploading to claude.ai would fail: the description is over 200 characters (215). Use "Let the AI install it" instead

Option 1Let Claude Code install it

Copy this sentence and paste it into a Claude Code chat

Where to install

Checking the commands · copy buttons appear once they pass

Checking the commands · copy buttons appear once they pass

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 /xhs-writer-skill 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.

  1. 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

  2. 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

    Checking the commands · copy buttons appear once they pass

  3. Restart Claude Code, then type /xhs-writer-skill or ask for a matching task and Claude picks it up

Skills you install yourself work in the ChatGPT desktop app, Codex CLI and the Codex IDE extension · ChatGPT on the web and mobile cannot use skills you install yourself, only skills that come with a plugin from the directory

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

Where to install

Checking the commands · copy buttons appear once they pass

Checking the commands · copy buttons appear once they pass

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 xhs-writer folder at ~/.agents/skills/xhs-writer

~ is your home folder ($HOME)

For Terminal users

A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.

  1. 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

  2. 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

    Checking the commands · copy buttons appear once they pass

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

Where to install

Checking the commands · copy buttons appear once they pass

Checking the commands · copy buttons appear once they pass

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.

  1. 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

  2. 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

    Checking the commands · copy buttons appear once they pass

  3. Restart 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

Where to install

Checking the commands · copy buttons appear once they pass

Checking the commands · copy buttons appear once they pass

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 xhs-writer-skill

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.

  1. 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

  2. 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

    Checking the commands · copy buttons appear once they pass

  3. In Cursor’s Agent chat, type / and pick xhs-writer-skill

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

Where to install

Checking the commands · copy buttons appear once they pass

Checking the commands · copy buttons appear once they pass

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.

View on GitHub ↗

Skill folders checked against the official Claude Code, Codex and Cursor guides on Nov 11, 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: xhs-writer
description: >-
  生成小红书(RedNote)笔记:把主题或用户提供的素材(文字/图片/视频)做成一套竖版卡片(图文)
  或一段带分镜脚本的视频稿(视频),配 caption + hashtags,落盘到 output/小红书/。
  使用时机:用户说 写小红书 / 做小红书笔记 / 小红书图文 / 小红书种草 / 出一套 xhs 卡片 /
  小红书视频 / write a Xiaohongshu note / make a RedNote post。
---

# xhs-writer — 小红书笔记生成

把一个主题 + (可选)用户素材,生成**图文卡片组**或**短视频脚本**,按规范落盘到 `output/小红书/`。

## 核心理念

小红书读者看的不是长文,是**卡片组**或**短视频**。长文原稿只是中间产物;caption 只是发布配文。

- **图文帖(image post)**:3-9 张 9:16 竖版卡片(cover + content × N + ending),每张 ≤80 字中文。
- **视频帖(video post)**:一段 15-90 秒竖屏视频脚本 + 封面卡,分镜写在 `meta.json.shots` 里,视频合成本 skill 不做。

### 爆款方法论(通用原则)

**5大核心原则**:
1. **真实素材优先**:项目截图、对比图、演示图 > 纯 AI 生图
2. **聚焦核心卖点**:用公式判断优先级(见 Step 0.5)
3. **素人感设计**:纯色背景 + 大 emoji + 口语化 > 品牌宣传风格
4. **痛点导向**:说"能解决什么问题" > 堆砌功能列表
5. **快速迭代**:V1(60分) → 用户反馈 → V2(70分) → 持续优化

**详细方法论**:见 `references/xiaohongshu-viral-methodology.md`

## 工作流

按顺序执行。**不要跳步**。

### Step 0 — Intake(必问,一次问完)

收到请求后不要动手,先向用户确认以下要点,尽量一条消息问完:

1. **主题 / 目标读者 / 核心观点**
2. **输出形态**:图文 or 视频?(默认图文)
3. **素材**:有没有已有的文字/图片/视频要用?贴路径或拖文件
4. **风格**:简约清新 / 科技感 / ins 风 / 商务 / 文艺复古 / 可爱卡通(默认"简约清新")
5. **卡片数量 / 视频时长**:图文默认 5-7 张;视频默认 30-60 秒

### Step 0.5 — 卖点/亮点分析(推荐执行)

**触发条件**:
- 推广类:产品、项目、工具、服务
- 分享类:好物推荐、经验总结、知识科普
- 测评类:产品对比、使用体验

**执行**:
1. 列出所有功能/特性/亮点
2. 用 AskUserQuestion 让用户打分(每个点):
   - 稀缺性(1-5):别人有吗?独特吗?
   - 实用性(1-5):能解决多大问题?
   - 可感知(1-5):用户能直接看到/感受到吗?
3. 计算得分 = 稀缺性 × 实用性 × 可感知
4. 排序,选择 Top 1-2 作为核心卖点/亮点

**输出**:
```markdown
## 卖点/亮点分析结果

| 点 | 稀缺性 | 实用性 | 可感知 | 得分 | 优先级 |
|---|---|---|---|---|---|
| 亮点A | 5 | 5 | 5 | 125 | 🥇 核心 |
| 亮点B | 2 | 3 | 3 | 18 | 🥉 辅助 |

**核心卖点/亮点**:亮点A(聚焦这个,其他作为辅助)
```

**参考**:`references/xiaohongshu-viral-methodology.md` 第2节

### Step 1 — 素材清点(有素材才做)

如果用户提供了素材路径,**先跑脚本生成清单**,再由 AI 用多模态能力填描述:

```bash
python3 scripts/analyze_material.py <path>... \
  --out <work-dir>/reference/materials.json \
  --frames-dir <work-dir>/reference/frames
```

脚本只做确定性预处理(分类、取分辨率/时长、抽帧)。AI 随后用视觉能力 **打开** `materials.json` 里每条 image/frame,填 `caption` 和 `usage`(cover / content-N / ending / reference)。详见 [`references/material-intake.md`](references/material-intake.md)。

**素材分类(按价值排序)**:
1. **对比图**(最有价值):before/after、input/output、不同版本对比
2. **功能演示**:界面截图、操作流程图、效果展示
3. **数据图表**:性能对比、用户增长、功能覆盖
4. **品牌素材**:Logo、配色方案、官方截图

**素材使用策略**:
- ✅ **优先级**:真实素材 > 图生图 > 代码叠加 > 纯 AI 生图
- ✅ **对比图**:用图生图(保留真实感 + 叠加小红书风格文字)
- ✅ **纯文字卡片**:用 AI 生图(素人感设计)
- ❌ **避免**:所有卡片都用纯 AI 生图(缺少真实感)

**参考**:`references/xiaohongshu-viral-methodology.md` 第3节

### Step 2 — 采集外部参考(观点类 / 资讯类必做;纯素材驱动可跳过)

按 [`references/reference-search.md`](references/reference-search.md) 执行,结果写入同一 `reference/` 目录。核心数据 ≥2 个来源交叉验证。

### Step 3 — 写长文原稿(2000-4000 字)

从 H2 开始(**不写 H1**;标题入 `meta.json.title`)。写完**先给用户看**,确认主旨再继续,不要直接跳到卡片/分镜。反模式清单以 [`references/humanizer-zh.md`](references/humanizer-zh.md) 为准。

### Step 4 — 去 AI 化

读 [`references/humanizer-zh.md`](references/humanizer-zh.md) 五层原则,完整扫描重写原稿,给出质量评分(满分 50)。

### Step 4.5 — 图生图处理(有真实素材时必做)

**触发条件**:Step 1 发现了对比图、功能演示等高价值素材

**工具**:gpt-image-2 图生图(需要 OpenAI API key)

**执行**:
```python
# 使用本项目的 image_generator
import sys
sys.path.insert(0, '~/.claude/skills/xhs-writer-skill/scripts')
from image_generator import GptImage2Generator

generator = GptImage2Generator(aspect_ratio="9:16")

# 图生图:保留真实素材 + 叠加小红书风格文字
generator.generate_scene_image(
    scene_data={
        'index': 1,
        'image_prompt': """Based on the reference image, create a Xiaohongshu style card, 9:16 vertical.

Keep the original image visible.

Add overlays:
- Top: "{标题}"
- Bottom: "{引导文字}"

Style:
- Keep original clear
- Casual Xiaohongshu style
- Authentic feel"""
    },
    output_path='output.jpg',
    size='auto',
    reference_image_path='material.jpg'
)
```

**Prompt 模板**:见 `references/xiaohongshu-viral-methodology.md` 附录C

**失败处理**:
1. 重试 1 次
2. 降低图片分辨率(max 1024px)
3. 改用代码叠加文字(`scripts/text_on_image.py`)

### Step 5 — 分发:图文 or 视频

#### 5A. 图文帖:拆成 3-9 张卡片

- 结构:`cover`(第 1 张) + `content`(中间若干) + `ending`(最后 1 张)
- 每张 ≤80 字(`title` + `content` 合计,代码点计数)
- 一张卡只讲一个论点;数字 / 对比 / 金句优先上卡
- 全套卡片 emoji 风格与配色保持一致

每张卡选一种 **合成策略**(写入 `cards[i].synthesis_strategy`),五选一。详见 [`references/material-intake.md`](references/material-intake.md):

| strategy | 适用 | 工具 | 优先级 |
|---|---|---|---|
| `img2img` | 有真实素材(对比图/演示图) | gpt-image-2 图生图 | 🥇 最佳 |
| `text_on_photo` | 有 1 张合适照片 + 一句钩子 | `scripts/text_on_image.py` | 🥈 次选 |
| `collage` | 有 2-4 张互补照片 | `scripts/collage_3x4.py` | 🥉 可用 |
| `pure_text` | 无素材,纯文字卡 | AI 生图(素人感) | ✅ 常用 |
| `ai_generated` | 概念图 / 数据图 | 用户自备 t2i 服务 | ⚠️ 慎用 |

**策略选择原则**:
- ✅ 有真实素材 → 优先 `img2img`(保留真实感)
- ✅ 纯文字卡片 → 用 `pure_text`(素人感设计)
- ❌ 避免所有卡片都用 `ai_generated`(广告感强)

素材有水印 → 先跑 `scripts/crop_watermark.py` 或按 [`references/image-sourcing.md`](references/image-sourcing.md) 处理。

#### 5B. 视频帖:写分镜脚本

- 写 6-12 个分镜,每镜 2-8 秒,累计时长对齐用户预期
- 每镜含:`narration`(口播,≤30 字)、`on_screen_text`(屏幕字,≤15 字)、`visual`(画面描述)、`material_ref`(若引用 `materials.json` 里某条素材)
- 仍要出一张 `cover` 卡(3:4)作为封面;视频本体由用户侧工具合成,本 skill 只产脚本

字段结构见 [`references/meta-schema.md`](references/meta-schema.md) 的 `shots[]` 定义。

### Step 6 — caption + hashtags + 标题

**标题生成(5种公式)**:
1. **痛点+解决方案**:`{具体痛点}?{解决方案}`
   - 适用:有明确痛点的工具/产品
2. **提问式**:`有没有那种{功能描述}的{产品类型}?`
   - 适用:新工具推荐、功能发现
3. **发现式**:`我发现了个宝藏!{核心价值}`
   - 适用:兴奋分享、好物推荐
4. **热点词**:`{热点词}爆火后,我用它做了{场景}`
   - 适用:蹭热点、技术类产品
5. **身份共鸣**:`{身份标签}必备!{核心功能}`
   - 适用:有明确目标人群的产品

**参考**:`references/xiaohongshu-viral-methodology.md` 附录F

**caption**:
- 100-300 字,hook 开头(数字/提问/惊叹) → 关键信息 → 行动号召(点赞/收藏/关注)
- 带 emoji,闺蜜语气
- **结构**:痛点共鸣 → 解决方案 → 具体功能 → 真实案例 → CTA
- **避免**:堆砌功能、正式文案、广告感

**hashtags**:
- 5-8 个,与主题强相关
- **核心标签**(4个):热点词 + 核心功能 + 差异化卖点 + 目标人群
- **辅助标签**(4个):场景词 + 品类词
- 避免 `#生活` 等过度泛化标签

只写进 `meta.json`,**不**粘进卡片或正文

### Step 7 — 落盘

目录与命名规则见 [`references/output-spec.md`](references/output-spec.md)。

```
output/小红书/{YYYY-MM-DD}/{短标题}_{YYYYMMDDHHmm}/
├── {完整标题}.md        # 长文原稿
├── meta.json            # 元数据(卡片 / 分镜 / caption / hashtags / materials)
├── images/              # (图文)最终卡图 / (视频)封面
└── reference/           # materials.json / 搜索结果 / summary / 思考过程
```

目录短标题与时间戳**必须**走脚本标准化,别手写:

```bash
python3 scripts/normalize_slug.py "原始长标题" --with-ts
```

`meta.json` 完整字段定义见 [`references/meta-schema.md`](references/meta-schema.md)。

### Step 8 — 校验(强制)

写完 `meta.json` 后必须跑:

```bash
python3 scripts/validate_meta.py <work-dir>/meta.json
```

非 0 退出 → 读报错修 `meta.json` 再跑,直到 clean。不要把未校验的产物交给用户。

## 不要做的事

- 不写 H1;标题只放 `meta.json.title`
- 不在正文末尾写"参考来源 / References";只落到 `reference/`
- 不在 `.md` 里留 `【插入图片:...】` 占位符;图片同步下载 + 引用相对路径
- 不跳过 Step 4(去 AI 化)和 Step 8(validate)
- 不自己手算目录名 / 时间戳,一律走 `normalize_slug.py`

---

## 工具依赖

### 必需工具
- Python 3.8+
- PIL (Pillow)

### 可选工具
- OpenAI API key(用于图生图,推荐)
  - 配置:在 `~/.claude/skills/xhs-writer-skill/.env` 填入 `OPENAI_API_KEY`
  - 没有 API key 也能用,会生成纯文字卡片

### 图生图使用
```python
# 添加路径
import sys
sys.path.insert(0, '~/.claude/skills/xhs-writer-skill/scripts')
from image_generator import GptImage2Generator

# 初始化
generator = GptImage2Generator(aspect_ratio="9:16")

# 生成
generator.generate_scene_image(
    scene_data={'index': 1, 'image_prompt': '...'},
    output_path='output.jpg',
    size='auto',
    reference_image_path='material.jpg'  # 图生图模式
)
```

---

## 参考资料

### 核心方法论
- **`references/xiaohongshu-viral-methodology.md`**:完整爆款方法论
  - 卖点优先级判断公式
  - 素材使用策略(优先级排序)
  - 爆款卡片结构(标准6-7张)
  - 文案公式库(5种标题公式)
  - 视觉风格指南(素人感 vs 精美设计)
  - 图生图 Prompt 模板(3种场景)
  - 常见错误清单(5大错误 + 解决方案)
  - 工具使用指南

### 其他参考
- `references/humanizer-zh.md`:去 AI 化原则
- `references/material-intake.md`:素材处理流程
- `references/image-sourcing.md`:图片来源处理
- `references/meta-schema.md`:元数据字段定义
- `references/output-spec.md`:输出目录规范

---

## 快速开始

### 场景1:推广产品/项目

```bash
# 用户说:"帮我推广这个项目 /path/to/project"

# Step 0.5: 卖点分析
# → 列出功能,让用户打分(稀缺性×实用性×可感知)
# → 选出核心卖点

# Step 1: 素材盘点
# → 扫描项目截图、对比图、演示图
# → 分类:对比图(最有价值)> 功能演示 > 其他

# Step 4.5: 图生图
# → 对比图用图生图(保留真实感 + 小红书风格)
# → 纯文字卡片用 AI 生图(素人感)

# Step 6: 生成标题(5个选项)
# → 痛点式、提问式、发现式、热点词、身份共鸣

# 输出:6-7张卡片 + caption + hashtags
```

### 场景2:好物分享/经验总结

```bash

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.

Sign in to post

From GitHub

GitHub stars
212
Maker
JuneYaooo
License
Apache-2.0
Last code change
Apr 25, 2026
SKILL.md files in the repo (including translated copies and templates)
1
Topics
claude-code-skills, xhs-skill, xhs-writer, xiaohongshu-skills