Yuwen Publish Precheck
yuwen-cool/yuwen-publish-precheck
国内自媒体发布前风险自审与保意修复:审口播稿、文章、图文笔记、字幕、封面文字能不能发抖音/小红书/微信视频号,给出具体位置、依据和可直接替换的改稿;被限流/删除/处罚后帮你复盘归因;你的行业敏感词、误报白名单和踩坑案例会沉淀成个人规则库,越用越准。Use when…
内容生产和分发统一管线。素材收集→出稿→排版→封面→朋友圈文案→多平台转换→一键分发。涵盖公众号写作、小红书轮播图、即刻文案、播客音频、品牌视频、Chrome CDP 自动发布。
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install OrangeViolin/content-pipeline content-pipeline --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
Claude Code skills documentation · loads skills from .claude/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .claude/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install OrangeViolin/content-pipeline content-pipeline --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .agents/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install OrangeViolin/content-pipeline content-pipeline --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .cursor/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install OrangeViolin/content-pipeline content-pipeline --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .gemini/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install OrangeViolin/content-pipeline content-pipelineInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .github/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install OrangeViolin/content-pipeline content-pipeline --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "content-pipeline" agent skill from https://github.com/OrangeViolin/content-pipeline/tree/main into .opencode/skills/content-pipeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "content-pipeline", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
content-pipeline内容生产和分发统一管线。素材收集→出稿→排版→封面→朋友圈文案→多平台转换→一键分发。涵盖公众号写作、小红书轮播图、即刻文案、播客音频、品牌视频、Chrome CDP 自动发布。
Content Pipeline is an agent skill from OrangeViolin/content-pipeline. 内容生产和分发统一管线。素材收集→出稿→排版→封面→朋友圈文案→多平台转换→一键分发。涵盖公众号写作、小红书轮播图、即刻文案、播客音频、品牌视频、Chrome CDP 自动发布。
Its SKILL.md is about 3.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 47 other files, including scripts and reference files (for example `README-EN.md`, `README.md` and `evolve/evolution-log.md`).
It sits in Productivity & Automation. It works with Xiaohongshu and WeChat. The repository describes itself as: AI-powered content production pipeline for creators. One prompt → multi-platform publishing. Claude Code Skill. The licence is MIT.
4 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 976283c. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
npxpython3From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use npx, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Content Pipeline loads about 3.9k tokens when it runs, and up to ~50k if it reads all its reference files. Until then it costs about 27 tokens; SKILL.md has 1,163 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check noted patterns worth knowing about, such as sudo or a known installer.
> `$MD_FORMATTER_DIR` 需在 `local/.env` 或环境变量中配置。> `$REMOTION_DIR` 需在 `local/.env` 或环境变量中配置。Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
The full file from OrangeViolin/content-pipeline at commit 976283c, republished under its MIT licence (© OrangeViolin). 1,163 words, ~3,883 tokens.
.claude/skills/content-pipeline/SKILL.md (or your agent's skills folder). This skill also uses 44 other files; get the full folder from GitHub.本技能默认使用 01fish 品牌色作为示例。你可以在
references/目录下修改为自己的品牌色,或在local/SKILL.local.md中覆盖路径和个人设定。
一条龙:素材收集 → 写文章 → 排版 → 多平台内容 → 一键分发。
[输出目录]/
├── drafts/
│ └── current.json # 当前素材列表
├── [文章标题].md # 产出的文章
├── [文章标题]_preview.html # 排版预览
└── [主题]-小红书版.html # 小红书轮播图输出目录由用户指定,或在 local/SKILL.local.md 中配置默认路径。未指定目录时,多平台产出放 /tmp/。
| 触发词 | 说明 |
|---|---|
/story | 查看当前素材状态 |
| "看看素材" | 查看已记录的素材 |
| "出稿" | 生成文章 + 排版 + 封面图 |
| "清空素材" | 清空当前素材 |
| "记一笔:xxx" | 手动添加素材 |
| "素材+1:xxx" | 手动添加素材 |
| "写个朋友圈" | 根据素材/文章生成朋友圈文案 |
| 触发词 | 说明 |
|---|---|
/xiaohongshu + 微信链接 | 微信文章转小红书轮播图 |
| "转小红书" + 微信链接 | 同上 |
| "做成小红书" + 微信链接 | 同上 |
| "转即刻" + 微信链接 | 生成即刻文案 |
| "转播客" + 微信链接 | 生成播客脚本 + AI 语音 |
/podcast + 链接/文章/书名 | 小宇宙播客全流程(15分钟百家讲坛风格) |
| "做播客" + 链接/文章/书名 | 同上 |
| "录播客" + 链接/文章/书名 | 同上 |
| "讲书播客" + 书名/链接 | 百家讲坛风格讲书播客 |
/shiji + 文章/素材 | 史记罗生门栏目播客(AI侦探×史源追踪) |
| "史记罗生门" + 文章/素材 | 同上 |
| "做史记播客" + 文章/素材 | 同上 |
| "做视频" + 微信链接 | 触发品牌视频管线 |
| "做视频画布" + 网址列表 | 生成录屏画布(全屏网页演示 + 露脸 + 提词器) |
| "录屏画布" + 网址列表 | 同上 |
| "录屏" + 网址列表 | 同上 |
| "多平台分发" + 微信链接 | 一次生成所有平台内容 |
| "转小红书并发布" + 微信链接 | 生成 + 自动触发分发 |
| 触发词 | 说明 |
|---|---|
/read-gzh + 微信链接 | 抓取并总结公众号文章 |
| "帮我读一下这篇公众号" | 同上 |
| "总结一下这篇文章" | 同上 |
| 触发词 | 说明 |
|---|---|
| "排版" | 用 01fish 主题排版 Markdown → 公众号 HTML |
| "做头图" / "封面图" | 生成公众号头图 HTML(浏览器下载 PNG) |
| "做竖版封面" / "竖版头图" | 从公众号封面 → 生成 3:4 竖版封面(1080×1440),适合小红书/视频号 |
| "做配图" / "准备配图" | 生成文章配图 HTML(浏览器下载 PNG) |
| "排版+配图" / "全套排版" | 排版 + 头图 + 配图一起生成 |
| 触发词 | 说明 |
|---|---|
/distribute | 读取 manifest 一键发布 |
| "一键发布" | 全平台发布 |
| "全平台发布" | 同上 |
| "发布到小红书" | 单平台发布 |
| "发布到即刻" | 单平台发布 |
AI 生成的内容必须诚实标注,不装人类,展现真实的创作过程。
当 AI 参与内容创作时,必须在文章中明确标注:
**调研 & 撰写**:AI(Claude)
**主导 & 审校**:[用户名]
**创作时间**:[实际用时](调研 X 分钟 + 写作 Y 分钟)❌ 不要写:
✅ 应该写:
边干活边记录 → 说"出稿" → 写文章 → 排版 → 封面图 → 朋友圈文案 → manifest适用场景:日常和 cc 协作时,自动积累素材,攒够了一键出稿。
微信链接 → 抓取文章 → 分析结构 → 生成小红书/即刻/播客/视频 → manifest → 分发适用场景:已有公众号文章,一键转为多平台内容并发布。
cc 在对话中主动识别有料瞬间并自动记录,无需手动触发。
识别信号:
| 类型 | 识别信号 | 示例 |
|---|---|---|
| 踩坑翻车 | 预期≠结果、报错、折腾半天 | "试了三种方案都不行" |
| 意外发现 | "没想到"、"原来可以"、意外有效 | "居然这样就解决了" |
| 迭代打磨 | 改了多版、从复杂到简洁 | "200行改成20行还能跑" |
| 搞笑时刻 | 对话金句、AI抽风、神奇bug | "它认真地给我写了一堆错的" |
| 突破时刻 | 卡了很久终于通 | "困扰一周的bug终于找到了" |
| 方法沉淀 | 可复用的技巧、心得 | "以后遇到这种情况就这么办" |
自动记录时:不打断对话,段落结尾标记 (✓ 素材+1)
用户说"记一笔:xxx"或"素材+1:xxx"时记录。
{
"topic": "主题(可选,出稿时自动提取)",
"materials": [
{
"time": "2026-01-30 14:30",
"content": "素材内容",
"type": "搞笑时刻",
"context": "可选的上下文备注",
"auto": true
}
],
"created": "2026-01-30"
}drafts/current.json| 内容类型 | 判断信号 | 使用框架 | 参考文件 |
|---|---|---|---|
| 说明书类 | 开源项目介绍、工具/产品说明、知识库/数据集发布、平台使用指南、"介绍一下 xxx" | 六段式说明书框架 | references/manual-framework.md |
| 教程类 | 教人安装/使用/配置工具、Skill 介绍、技术实战、"怎么做 xxx" | 六段式教程框架 | references/tutorial-framework.md |
| 深度长文 | 行业分析、人物故事、趋势判断、观点输出、"为什么 xxx" | 四幕式深度框架 | references/writing-style.md |
说明书类文章框架(3000-6000 字):
项目定义+核心数据(标题即摘要,开头即高潮)
→ 一、核心成果(数据总览表)
→ 二、功能特性(逐一展开,每个一小节+配图)
→ 三、怎么用(按用户分层:零门槛→进阶→开发者→创作者)
→ 四、价值/洞见(超出工具本身的意义)
→ 五、扩展路线(可选)
→ 写在最后(核心价值+愿景+链接)教程类文章框架(2000-4000 字):
先看结果(截图+成品+链接)
→ 一、核心概念是什么(表格+一句话定义)
→ 二、怎么安装/使用(分步骤+代码块+配图标记)
→ 三、实战演示(分阶段+表格展示+人机协作)
→ 四、拿走即用(快速安装命令+使用方式表格)
→ 写在最后(升华+CTA)深度长文框架(8000-12000 字):
序言(故事先行,700 字不出论点)
→ 01 铺设背景
→ 02 核心论述
→ 03 转折/案例
→ 04 升华/收束references/writing-style.md。这是从鱼头头 9 篇已发布文章中逆向工程出的真实写作模式,不是理论指南。核心要点:极短段落(1-3句)、单句成段做"钉子"、口语化动词、"不是A是B"金句句式、数字制造反差、表格优先于段落。references/cover-template.md)references/cover-vertical-spec.md_preview.html 的对应位置插入占位符注释 <!-- IMAGE:配图-1.png -->,分发脚本会自动将配图上传到微信 CDN 并替换占位符为 <img> 标签。references/platform-copy.md)/distribute 使用。WeChat 发布流程(全自动):
md2wechat_formatter.py 生成 _preview.html)_preview.html 中插入图片占位符 <!-- IMAGE:文件名.png -->/distribute --platforms wechat(API 自动:上传封面→上传配图→替换占位符→推送草稿箱)wechat 部分字段说明:
wechat.html(必填):排版后的 _preview.html 路径(配图位置用 <!-- IMAGE:文件名.png --> 标记)wechat.cover_image(必填):封面 PNG 路径wechat.title(必填):文章标题wechat.author:作者名(默认 01fish)wechat.digest:文章摘要(120 字内)wechat.images:配图 PNG 路径列表(按文章中出现的顺序排列,分发脚本会上传并插入到对应占位符位置)wechat.markdown(可选):文章 Markdown 路径(仅作记录,不再用于转换)cd "$MD_FORMATTER_DIR"
python3 md2wechat_formatter.py [文章路径] --theme [主题] --font-size [字号]
$MD_FORMATTER_DIR需在local/.env或环境变量中配置。
推荐主题:01fish(01fish 品牌色,默认)、chinese(中国风)、apple(极简优雅)
推荐字号:medium(15px 默认)、large(16px 长文推荐)
01fish 主题说明:基于 chinese 主题,使用 01fish 品牌色(墨绿 #1A3328 + 鱼红 #C44536 + 宣纸底 #F2EDE3)
使用 Python 抓取脚本(微信有反爬验证,WebFetch 会被拦):
python3 "${SKILL_DIR}/scripts/fetch_wechat_article.py" "<URL>" --json超时 30 秒。失败则提示用户手动复制文章正文。
如果用户只是说"帮我读一下这篇公众号"(/read-gzh 触发),执行抓取后直接生成结构化总结,不进入后续内容生成流程。总结格式:
# 文章总结
## 基本信息(标题/作者/类型/配图数)
## 核心观点(3条)
## 关键信息
## 金句摘录
## 图片内容(下载并识别配图中的文字)
## 思考/迭代点提取:标题、副标题/金句、核心概念、关键数据、步骤/流程、亮点/特色、方法论/金句、行动召唤。
8-10 张卡片,遵循小红书阅读节奏(→ 读 references/xiaohongshu-format.md):
| 位置 | 卡片类型 | 内容 |
|---|---|---|
| 第 1 张 | 封面 | 大标题 + hook + 迷你视觉元素 |
| 第 2 张 | 先看结果 | 成品展示 + 核心数据 |
| 第 3-4 张 | 概念解释 | 核心概念拆解 |
| 第 5-7 张 | 流程/实战 | 步骤、对比、流程图 |
| 第 8 张 | 亮点/特色 | 产品/作品亮点卡片 |
| 第 9 张 | 方法论 | 一句话金句提炼 |
| 第 10 张 | 行动召唤 | 链接 + 社区引导 |
输出路径:文章同目录下 [简短主题]-小红书版.html,未指定目录放 /tmp/。浏览器自动打开预览。
最后一张行动召唤页必须包含:微信号 [你的微信号](强调色大字)、备注关键词、核心链接。
在
local/SKILL.local.md中配置你的实际微信号。
📚 重要:生成前必读范例
参考 references/xiaohongshu-examples/观鸟图鉴-范例.html 的质量标准:
✅ 卡片设计要求
✅ 文案质量要求
生成的内容应达到范例的专业水准。
根据内容类型选择风格:
个人 IP 风格(真人分享、产品开发、踩坑记录)
local/SKILL.local.md 中指定的个人品牌风格文件(如有)local/SKILL.local.md 中自定义01fish风格(方法论总结、深度分析)
references/platform-copy.md 的小红书部分→ 读 references/platform-copy.md 的即刻部分。
根据触发词选择播客模式:
| 触发词 | 模式 | 时长 | 风格 |
|---|---|---|---|
| "转播客" | 标准模式 | 5-8 分钟 | AI搭档聊天风 → 读 references/platform-copy.md 播客部分 |
/podcast / "做播客" / "录播客" / "讲书播客" | 百家讲坛模式 | 15 分钟 | 讲书人风格 → 读 references/xiaoyuzhou-podcast.md |
百家讲坛模式是小宇宙播客的主力模式,适合把文章/书籍改编为有故事感、有节奏感的深度音频内容。
通过 SSH 连接 Ubuntu 机器,使用 IndexTTS2 本地生成(零样本声音克隆)(→ 读 references/tts-config.md)。
百家讲坛模式的 TTS 调整:15 分钟脚本约 4000 字,分段大小 600 字(更短分段带来更好的语音节奏),预计 6-7 段。
文件命名:[播客标题].mp3 + [播客标题]-播客脚本.md
→ 读 references/podcast-cover-template.md
生成 3000×3000 正方形封面 HTML,01fish 墨绿体系,浏览器下载 PNG 后上传小宇宙。
文件命名:[播客标题]-播客封面.html
所有内容生成完毕后,自动输出 manifest.json 到输出目录。格式:
{
"version": "1.0",
"created": "<ISO时间戳>",
"source": "<微信链接>",
"title": "<文章标题>",
"outputs": {
"xiaohongshu": { "html": "...", "copy": { "title": "...", "body": "...", "tags": [...] } },
"jike": { "copy": { "body": "...", "circles": [...] } },
"xiaoyuzhou": { "audio": "...", "script": "...", "cover": "...", "copy": { "title": "EP01丨...", "description": "...", "show_notes": "..." } },
"video_canvas": { "html": "...", "teleprompter_md": "...", "cover_html": "..." }
}
}如果用户说"转小红书并发布",生成 manifest 后自动执行 /distribute。
仅当用户提到"视频"、"抖音"、"视频号"或"品牌视频"时执行:
A. Remotion 品牌片头片尾
cd "$REMOTION_DIR"
npx remotion render src/index.ts Intro --output /tmp/brand-intro.mp4
npx remotion render src/index.ts Outro --output /tmp/brand-outro.mp4
$REMOTION_DIR需在local/.env或环境变量中配置。
B. AI 视频 Prompt — 为 Seedance 2.0 或 Google Veo 生成 4 段视频 prompt
C. ffmpeg 拼接指令 — 生成拼接命令供用户手动执行
仅当用户说"做视频画布"、"录屏画布"、"录屏"时执行。用户提供要演示的网址列表 + 简短主题。
[提示] cue 标记[主题]-提词器脚本.md,用户可直接编辑references/video-canvas-template.md 获取完整 CSS+JS 模板,网址预填 WEB_URLS + 提词器脚本填入 SCRIPTS[主题]-视频画布.html,保存到用户指定目录或 /tmp/[主题]-封面.html,暗底 + 人像圆框,浏览器下载 PNG告知用户所有产出物路径,提示可调整,输入 /distribute 可一键发布。
公众号同步提示:封面 PNG 从浏览器下载后,直接 /distribute --platforms wechat 即可同步到草稿箱(API 模式,无需打开 Chrome)。
一次性产出五样东西,不需要额外要求:
/distribute 一键发布)单独触发 /podcast 时,产出三样:
第 9B 步可选追加(说"视频画布"/"录屏"时): 6. 录屏画布 HTML(全屏网页演示 + 摄像头露脸 + 录制 + 提词器 + 美颜,16:9 固定) 7. 提词器脚本 md(按网站分段,可编辑,修改后说"更新提词器"同步到 HTML) 8. 封面图 HTML(暗底 + 人像圆框,浏览器下载 PNG)
读取 manifest.json,通过 Chrome CDP 自动化发布到各平台(→ 读 references/distribute-platforms.md)。
# 全平台发布
npx -y bun "${SKILL_DIR}/scripts/distribute/distribute.ts" --manifest /path/to/manifest.json
# 选择平台
npx -y bun "${SKILL_DIR}/scripts/distribute/distribute.ts" --manifest /path/to/manifest.json --platforms xhs,jike
# 预览模式(不提交,只预填内容)
npx -y bun "${SKILL_DIR}/scripts/distribute/distribute.ts" --manifest /path/to/manifest.json --platforms xhs --preview| 缩写 | 平台 | 状态 |
|---|---|---|
wechat | 公众号 | 可用 |
xhs | 小红书 | 可用 |
jike | 即刻 | 可用 |
xiaoyuzhou | 小宇宙 | 可用 |
douyin | 抖音 | 实验性 |
shipinhao | 视频号 | 待开发 |
公众号 → 小红书 → 即刻 → 小宇宙 → 抖音 → 视频号(顺序执行,避免 Chrome 端口冲突)
| 级别 | 模式 | 触发条件 |
|---|---|---|
| L0 | API 直推 | 公众号 API 直接推草稿箱,无需 Chrome |
| L1 | 自动发布 | CDP 完全自动化 |
| L2 | 辅助发布 | 登录态失效/选择器失效/--preview |
| L3 | 手动模式 | CDP 连接失败 |
公众号优先 L0(API),凭证缺失或失败时自动降级 L3(手动)。
凭证来源(按优先级):
WECHAT_APPID + WECHAT_APPSECRET~/.config/wechat-api/config.jsonToken 缓存:
~/.config/wechat-api/token-cache.json执行流程:
_preview.html 提取文章内容(extractArticleContent)fixHtmlForWechat)uploadLocalImagesInHtml)thumb_media_iddraft/add 创建草稿两套品牌色体系:
单一真相源:在
local/SKILL.local.md中指定你的品牌色文档路径。 如果色值冲突,以品牌文档为准。以下色板作为默认示例。
比例法则:墨绿 85% : 鱼红 5% : 其余 10%
| 名称 | 色值 | 用途 |
|---|---|---|
| 墨绿主色 | #1A3328 | 暗底卡片背景 |
| 宣纸底 | #F2EDE3 | 浅底卡片背景 |
| 鱼红 | #C44536 | 强调色、数字、标签(仅点睛) |
| 半透白 | rgba(255,255,255,0.5) | 暗底上的品牌名 |
| 半透墨绿 | rgba(26,51,40,0.4) | 浅底上的品牌名 |
| 苔灰 | #7A8C80 | 次要文字 |
| 深墨 | #0F1F18 | 更深背景 |
| 淡青 | #D4DDD7 | 分割线、边框 |
比例法则:桃气粉 15% : 奶油黄 40% : 暮光紫 10% : 灰色 35%
| 名称 | 色值 | 用途 |
|---|---|---|
| 桃气粉 | #FF6B9D | 主强调色、标题、关键数据 |
| 奶油黄底 | #FFF9E6 | 浅底背景、卡片底色 |
| 暮光紫 | #9D7BA8 | 辅助色、次要信息、品牌名 |
| 温灰 | #6B6B6B | 正文文字 |
| 浅灰底 | #F5F5F5 | 现代感背景 |
| 深夜蓝 | #2D3047 | 暗底背景(少用) |
| 内容类型 | 使用品牌 | 原因 |
|---|---|---|
| 公众号深度文章 | 01fish | 专业、权威、内容品牌 |
| 行业分析报告 | 01fish | 冷静客观 |
| 小红书真人分享 | 鱼头头 | 温暖、真实、真人IP |
| 即刻日常动态 | 鱼头头 | 活泼、亲和 |
| 产品开发记录 | 鱼头头 | 真实过程展示 |
| B端产品介绍 | 01fish | 专业可信赖 |
双品牌联动:同一篇内容,公众号用01fish色,小红书转发用鱼头头色
font-family: -apple-system, 'PingFang SC', 'Microsoft YaHei', sans-serif;每页左上角 01fish logo + 文字,暗底页用 .light,浅底页用 .dark。
右下角 1/N,暗底页 rgba(255,255,255,0.2),浅底页 rgba(26,51,40,0.2)。
微信 → 小红书不是照搬,需适配:
| 维度 | 微信 | 小红书 |
|---|---|---|
| 篇幅 | 2000-3000 字 | 每页 50-80 字 |
| 结构 | 线性阅读 | 卡片式跳读 |
| 语气 | 技术向、深度 | 简洁、直观、有冲击力 |
| 视觉 | 文字为主 | 视觉为主、文字点缀 |
改写要点:标题要炸、数字要大、一页一个点、视觉替代文字、保留核心链接。
首次生成小红书图片时,参考 references/xiaohongshu-examples/ 目录下的范例文件获取完整 CSS + JS。
如果有额外的本地模板参考,在
local/SKILL.local.md中指定路径。
生成新内容时复用范例文件的 CSS + JS 部分,只替换卡片内容。
Agent Execution: Determine this SKILL.md directory as SKILL_DIR, then use ${SKILL_DIR}/scripts/<name>.
| Script | Purpose |
|---|---|
scripts/fetch_wechat_article.py | 微信文章抓取(Python,模拟微信 UA) |
scripts/distribute/distribute.ts | 分发主编排器 |
scripts/distribute/cdp-utils.ts | 共享 CDP 工具 + Manifest 类型定义 |
scripts/distribute/wechat-api.ts | 公众号 API 客户端(token 管理、图片上传、草稿创建) |
scripts/distribute/platforms/*.ts | 各平台发布模块 |
cc 按需读取,不要一次性加载所有 reference。
| 场景 | 读取文件 |
|---|---|
| 出稿写说明书文章 | references/manual-framework.md — 六段式说明书框架(定义+成果→特性→用法→价值→路线→收束,3000-6000 字) |
| 出稿写深度长文 | references/writing-style.md — 人设 + 写作规范 + 格式(四幕式,8000-12000 字) |
| 出稿写教程文章 | references/tutorial-framework.md — 六段式教程框架(先看结果→概念→操作→实战→拿走即用,2000-4000 字) |
| 生成头图/配图 | references/cover-template.md — 01fish 风格排版规范(头图 + 配图 + 视觉组件) |
| 横版→竖版封面 | references/cover-vertical-spec.md — 公众号封面转竖版的 CSS 转换规范 |
| 生成小红书轮播图 | references/xiaohongshu-format.md — HTML 模板 + 视觉组件库 |
| 生成各平台文案 | references/platform-copy.md — 小红书/即刻/播客/朋友圈文案规范 |
| 生成播客音频 | references/tts-config.md — IndexTTS2 本地 TTS 配置 + 生成脚本 |
| 小宇宙播客(百家讲坛) | references/xiaoyuzhou-podcast.md — 15分钟讲书人风格脚本规范 + 改编流程 + 发布配置 |
| 播客节目封面 | references/podcast-cover-template.md — 3000×3000 正方形封面 HTML 模板 |
| 史记罗生门栏目 | references/shiji-luoshengmen.md — 栏目品牌设定 + AI侦探风格 + 脚本结构 |
| 分发到各平台 | references/distribute-platforms.md — 平台配置 + manifest 格式 + 降级策略 |
| 生成录屏画布 | references/video-canvas-template.md — 录屏画布模板(全屏网页演示+露脸+提词器+录制) |
| 问题 | 处理 |
|---|---|
| 微信抓取失败 | 提示用户手动复制文章正文 |
| 文章太短(<500字) | 压缩为 5-6 张卡片 |
| 文章太长(>5000字) | 精选核心,控制 10 张以内 |
| 导出图片模糊 | 检查 SCALE=2,浏览器缩放 100% |
| manifest 不存在 | 提示先运行内容生成 |
| Chrome 启动失败 | 降级 L3(手动模式) |
| IndexTTS2 模型加载失败 | 检查 checkpoints 目录和 infer_v2 导入 |
| TTS 生成失败 | 只输出脚本文本,提示手动录制 |
© OrangeViolin, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 44 other files (scripts, references) in the repository root of OrangeViolin/content-pipeline.
Open the folder on GitHubat commit 976283c
Content Pipeline next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Content Pipeline this skillOrangeViolin/content-pipeline | 219 | — | ~3.9k | Automated safety check: Notes | MIT | |
| Yuwen Publish Precheckyuwen-cool/yuwen-publish-precheck | 775 | 1 repos | ~1.8k | Automated safety check: Pass | MIT | |
| Wechat Graphic Monetizationchenjin-cmd/wechat-graphic-monetization | 169 | — | ~798 | Automated safety check: Pass | MIT | |
| Feedgrab BatchiBigQiang/feedgrab | 614 | — | ~1.8k | Automated safety check: Pass | MIT | |
| Private Domainvivy-yi/xiaohongshu-skills | 478 | — | ~3.9k | Automated safety check: Pass | None | |
| Account Integrationvivy-yi/xiaohongshu-skills | 478 | — | ~7k | Automated safety check: Pass | None |
yuwen-cool/yuwen-publish-precheck
国内自媒体发布前风险自审与保意修复:审口播稿、文章、图文笔记、字幕、封面文字能不能发抖音/小红书/微信视频号,给出具体位置、依据和可直接替换的改稿;被限流/删除/处罚后帮你复盘归因;你的行业敏感词、误报白名单和踩坑案例会沉淀成个人规则库,越用越准。Use when…
chenjin-cmd/wechat-graphic-monetization
This skill should be used when the user wants to monetize WeChat "图文" (photo-text / 小绿书, the WeChat Official Account image-note format modeled on Xiaohongshu) via 达人带货 (affiliate-style product…
iBigQiang/feedgrab
Batch content grabber — bulk fetch bookmarks, user tweets, search results, author notes, wiki pages, subreddit posts, and more from X/Twitter, Xiaohongshu, Reddit, WeChat, YouTube, Feishu, Zhihu.
vivy-yi/xiaohongshu-skills
A skill your agent uses when driving Xiaohongshu followers to private domains like WeChat, building offline communities, implementing follower retention strategies beyond platform, or creating…
vivy-yi/xiaohongshu-skills
A skill your agent uses when managing multiple social platforms, cross-promoting content, or building omnichannel presence
LeoYeAI/openclaw-master-skills
Chinese multi-platform content matrix generator — given a topic, auto-generate content adapted for Xiaohongshu, WeChat Official Account, Douyin, and Bilibili with true style transfer (not just…
Works with
Categories
内容生产和分发统一管线。素材收集→出稿→排版→封面→朋友圈文案→多平台转换→一键分发。涵盖公众号写作、小红书轮播图、即刻文案、播客音频、品牌视频、Chrome CDP 自动发布。. Content Pipeline is an agent skill from OrangeViolin/content-pipeline.
Content Pipeline fits situations like: productivity & Automation work in your project.
Run `npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a claude-code`. Or copy the skill folder (the OrangeViolin/content-pipeline repository) into .claude/skills/content-pipeline in your project. Claude Code loads it when a task matches its description.
Run `npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a codex`. Or copy the skill folder (the OrangeViolin/content-pipeline repository) into .agents/skills/content-pipeline in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add OrangeViolin/content-pipeline --skill content-pipeline -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/content-pipeline, .gemini/skills/content-pipeline, .github/skills/content-pipeline and .opencode/skills/content-pipeline in your project.
Going by SKILL.md and its folder, Content Pipeline needs the command-line tools its instructions call (npx and python3). Our summary lists: Python 3.
SKILL.md contains no URLs. Its commands use npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
Content Pipeline is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
About 3.9k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 46k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Content Pipeline: Yuwen Publish Precheck (yuwen-cool/yuwen-publish-precheck, 775 stars), Wechat Graphic Monetization (chenjin-cmd/wechat-graphic-monetization, 169 stars), Feedgrab Batch (iBigQiang/feedgrab, 614 stars) and Private Domain (vivy-yi/xiaohongshu-skills, 478 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
OrangeViolin (a GitHub user) maintains it in OrangeViolin/content-pipeline, which has 219 GitHub stars. The repository was last updated on April 12, 2026.
Source: OrangeViolin/content-pipeline on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.