Xiaohongshu Ad Ops

Cleanup Review

python scripts/_purge_cache.py is an optional cleanup helper for the Xiaohongshu Ad Ops skill. It may recursively remove /tmp/skill_build/*, /tmp/skill_build/.cache. Show the exact paths first, explain whether they are skill-private temporary paths or user state, and run the cleanup only after the user confirms.

Xiaohongshu Ad Ops

Use this skill to turn a fuzzy ad idea into a practical Xiaohongshu ad workflow.

Default goal:

• clarify the offer
• define the campaign objective
• generate note/ad angles
• structure tests and budget
• improve lead quality and conversion

What this skill covers

• 小红书信息流 / 笔记广告 planning
• 种草 + 转化 combined workflows
• leads for services, training, local business, AI services, consulting, SaaS, ecommerce
• creatives, hooks, note structure, CTA, landing-page alignment
• testing logic and iteration loops

Core workflow

1. Clarify the business first Identify:

   • product / service
   • price band
   • target user
   • sales path
   • conversion action: 私信 / 表单 / 加企微 / 下单 / 留资
   • whether the goal is branding, lead gen, or direct conversion

2. Choose the ad mode

   • Lead-gen mode: 留资 / 私信 / 企业微信
   • Seeding mode: build认知 and搜索心智 first
   • Conversion mode: pull direct action with stronger CTA
   • Hybrid mode:种草 + 承接页 / 私域 / 销售转化

3. Build the offer frame Always answer:

   • why click now
   • why trust you
   • who this is for
   • what problem it solves
   • what happens after the click

4. Design the campaign structure Return when relevant:

   • campaign goal
   • audience buckets
   • angle matrix
   • creative/note directions
   • budget split
   • KPI targets
   • iteration rules

5. Write ad-ready outputs Prioritize:

   • hook lines
   • note titles
   • note outlines
   • CTA variants
   • landing-page /私信承接 copy
   • sales follow-up suggestions

Recommended output format

1. 投放目标

• 这轮广告到底要拿什么结果

2. 用户分层

• 冷用户
• 半热用户
• 高意向用户

3. 选题 / 角度矩阵

• 痛点角度
• 结果角度
• 案例角度
• 反常识角度
• 对比角度

4. 素材 / 笔记方向

• 标题
• 开头钩子
• 正文框架
• CTA

5. 预算与测试计划

• 先测什么
• 每组预算怎么分
• 观察哪些指标
• 什么时候关停 / 放量

6. 承接设计

• 落地页重点
• 私信首句
• 销售跟进脚本方向

What usually goes wrong

Avoid these mistakes:

• 只讲产品，不讲用户情境
• 只做种草，不设计承接
• 只追点击，不看留资质量
• 创意全是同一种角度
• 没有测试节奏和停放规则
• 笔记像广告话术，不像真实内容

Chinese ad-writing rules

• 开头先打情境或痛点，不先自夸
• 小红书内容要像内容，不像硬广
• 强调“适合谁 / 不适合谁”会更容易筛选线索
• 先让用户产生代入，再谈产品
• CTA 不要一味硬推，优先“先私信 / 先领取 / 先测一下”这类低门槛动作

Best use cases

This skill is especially strong for:

• AI 安装 / 培训 / 顾问服务
• 本地商家获客
• 高客单服务型产品
• 小团队投流测试
• 想把种草内容和销售承接串起来的业务

Bundled resources

• references/xhs-ad-principles.md — platform-fit ad logic and投放 thinking
• references/creative-matrix.md — angle matrix for ad/note ideation
• references/lead-gen-playbook.md — lead-generation and conversion handoff patterns
• references/note-templates.md — 爆款标题 / 笔记模板 / CTA 模板
• references/industry-templates.md — AI服务 / 本地商家 / 教育培训 / 高客单服务行业模板
• references/landing-and-dm.md — 落地页与私信承接模板
• references/review-template.md — 投放复盘模板
• scripts/xhs_ad_plan_brief.py — generate a structured Xiaohongshu ad plan brief from raw business info

Script usage

python3 scripts/xhs_ad_plan_brief.py <input.txt>

Use the script for a quick first-pass plan, then refine with business judgment.

Default stance

Small-budget practical. Less theory, more plan. Focus on conversion path, not vanity metrics.