Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add lindixu6-hash/ai-content-workflow-skills --skill xiaohongshu-viral-directorgit clone --depth 1 https://github.com/lindixu6-hash/ai-content-workflow-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lindixu6-hash/ai-content-workflow-skills/xiaohongshu-viral-director)<a href="https://agentmods.dev/skills/lindixu6-hash/ai-content-workflow-skills/xiaohongshu-viral-director"><img src="https://agentmods.dev/badge/skills/lindixu6-hash/ai-content-workflow-skills/xiaohongshu-viral-director/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/lindixu6-hash/ai-content-workflow-skills/xiaohongshu-viral-director"><img src="https://agentmods.dev/badge/skills/lindixu6-hash/ai-content-workflow-skills/xiaohongshu-viral-director.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00053 | $0.01775 |
| Opus 5 | $0.00026 | $0.00888 |
| Sonnet 5 | $0.00011 | $0.00355 |
| Haiku 4.5 | $0.00005 | $0.00178 |
Grade A, and why
xiaohongshu-viral-director scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
小红书爆款编导 Skill
目标:让用户在信息流里点击,在笔记里读完/滑完、收藏,并能在未来搜索时再次找到。
本 Skill 只负责编导决策,不代写完整正文。
在 Content OS “成长(给初稿)”节点内部执行:
小红书编导报告 → 写作者初稿
一、强制读取
每次调用必须完整读取:
references/lv-bai-xhs-benchmark.md
只学习公开方法,不复制标杆的标题、笔记和账号身份。
二、小红书平台逻辑
小红书同时有两条流量路径:
推荐流
封面/标题点击 → 首屏停留 → 滑动/阅读 → 收藏评论
搜索流
明确问题 → 长尾词匹配 → 内容解决问题 → 收藏/复访/决策
小红书编导优先级:
搜索意图 → 封面标题点击 → 看见即能用 → 收藏资产 → 真实体验 → 长尾复访
不是抖音的逐秒完播逻辑。
三、吕白式编导决策
1. 禁止自我陶醉
先写:
用户为什么要搜/点这篇,而不是作者为什么想表达?
作者经历只负责证明,不负责占满正文。
2. 爆款母题重复
创作前检查:
- 同搜索词下的高表现选题反复出现什么?
- 哪种封面信息一眼能读懂?
- Top评论还在追问什么?
- 本篇只升级哪个变量?
不照抄内容,重复经过验证的用户问题。
3. 专一
一篇只服务一个:
- 人群。
- 场景。
- 痛点。
- 搜索意图。
4. 给收藏理由
笔记必须留下一个以后能用的资产:
- 清单。
- 判断标准。
- 步骤。
- 对比。
- 模板。
- 截图卡。
- 路线/资源。
纯情绪帖也要提供一句可转述的身份表达。
四、输入
必须读取:
- 用户选题。
- 当前选中的芽。
- 当前选中芽的粉色标记。
- 用户真实经历。
- 图片与截图素材。
- 目标搜索词或用户问题。
- 账号定位。
若没有目标搜索意图,先从选题拆出:
人群 + 场景 + 痛点/任务
五、编导步骤
Step 1|确定搜索意图
选择一种:
- 想知道。
- 想学会。
- 想比较。
- 想避坑。
- 想做决定。
- 想表达自己的感受。
Step 2|确定笔记承诺
一句话写清:
点开并收藏后,用户以后能用它完成什么?
Step 3|选择内容形态
- 图文卡片:清单、步骤、判断、对比。
- 真实照片+长文:经历、身份、情绪表达。
- 视频:需要动作、过程、现场证明。
- Live Photo/混剪:材料本身有运动价值。
不因平台扶持盲目选视频。
Step 4|封面设计
封面只负责一个信息:
- 谁的问题。
- 什么场景。
- 具体结果/冲突。
要求:
- 手机缩略图可读。
- 1–2行主文案。
- 真实素材优先。
- 不把正文摘要塞满封面。
Step 5|标题设计
标题必须:
- 20字以内优先。
- 自然包含核心搜索词。
- 明确人群/场景/痛点。
- 给点击理由,不把答案全部讲完。
禁止关键词堆砌和抖音式空悬念。
Step 6|首屏与正文
首屏100字内:
- 复述用户正在经历的具体问题。
- 给核心结论或结果。
- 告诉用户这篇能拿走什么。
正文:
真实场景 → 证据 → 机制 → 可收藏资产 → 个人边界
Step 7|卡片结构
图文笔记每页只完成一项任务:
- 封面。
- 问题/结果。
- 证据。
- 方法或判断。
- 示例。
- 总结卡。
页与页之间有继续滑动的理由。
Step 8|搜索与评论
报告必须给出:
- 核心词。
- 2–4个长尾词。
- 评论区自然问题。
- 后续系列词。
六、小红书100分评分
| 维度 | 分值 |
|---|---|
| 搜索意图清晰 | 20 |
| 封面标题点击 | 20 |
| 收藏资产 | 20 |
| 真实证据 | 15 |
| 图文可扫读 | 15 |
| 评论与系列长尾 | 10 |
评级:
- 80–100:高潜力。
- 65–79:中高潜力。
- 50–64:中潜力。
- 低于50:先补搜索意图或资产。
不预测具体点赞、收藏和粉丝。
七、固定输出
# 小红书编导爆款预判报告
## 1. 结论
- 总分:
- 潜力:
- 目标搜索意图:
- 提升一级最少补什么:
## 2. 用户问题
- 人群:
- 场景:
- 痛点/任务:
- 点开后要完成的决定:
## 3. 笔记承诺
- 核心结果:
- 收藏资产:
- 作者经历承担什么证据:
## 4. 形态
- 图文/视频/Live Photo:
- 为什么:
- 素材缺口:
## 5. 封面与标题
- 封面主文案:
- 封面画面:
- 标题:
- 核心词:
- 长尾词:
## 6. 首屏
- 第一句:
- 前100字任务:
- 最高价值:
## 7. 页序/正文结构
| 页/段 | 任务 | 内容 | 证据 | 下一页理由 |
|---|---|---|---|---|
## 8. 搜索与传播
- 收藏理由:
- 谁会转发给谁:
- 评论触发:
- 后续系列:
## 9. 失败预演
- 最可能不点开的原因:
- 最可能不收藏的原因:
- 必须删除:
- 必须讲透:
## 10. 写作者交接
- 主线:
- 必须保留:
- 关键词位置:
- 结尾落点:
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 304 lines · 53 tokens per session scan A 994306881a72
xiaohongshu-viral-director is a skill published in the GitHub repository lindixu6-hash/ai-content-workflow-skills (4 stars, last pushed 12d ago), licensed MIT. It adds 53 tokens to every session and 1,775 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
sfumatoai-writing-skill
Turn keywords, long text, articles, URLs, files, screenshots, or mixed media into verified Xiaohongshu knowledge-explainer image posts for AI-interested beginners. Use when Codex must analyze the input instead of blindly summarizing it, choose the strongest topic, research ambiguous or technical claims with…
cyxj-blog-pub
A publishing workflow for adding a finished article to an Astro blog, a website built with the Astro framework.
cyxj-transcript
A workflow that turns a video or audio transcript into a structured article draft in Obsidian Markdown.
create-xiaohongshu-knowledge-carousel
Create complete Chinese Xiaohongshu educational carousel images from a topic, source document, notes, or an existing page concept. Use when Codex must plan every definition and meaning without compressing knowledge points, maintain a recurring character and visual system, directly render fully typeset 3:4 final pages…
cyxj-video-cover
A generator for making 16:9 video-cover background scenes without text, letters, numbers, logos, or watermarks.
multi-wordcheck
A prohibited-word checker for content published on WeChat public accounts, Xiaohongshu and Douyin. It checks text, files, images and links against official word lists and marks matches with replacement suggestions in context.