Xiaohongshu topic planner
Skill ShenyuanNext/wepr-growth-skills/skills/xiaohongshu-topic-planner
Build evidence-based Xiaohongshu topic systems, series, backlogs, and 7/14/30-day calendars from account positioning, user situations, search intent, customer questions, objections, proof, production capacity, and business goals. Use for 小红书选题, 内容方向, 系列策划, 周更计划, 选题池, 内容日历, 产品种草选题, 服务号规划, or personal-IP topic planning. Route title-only requests to xiaohongshu-title and final-copy requests to plan-xiaohongshu-growth.From its SKILL.md
npx -y skills add ShenyuanNext/wepr-growth-skills --skill xiaohongshu-topic-plannerAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 28 days oldThe repository was created 28 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
3.8 KB, 665 tokens by cl100k_base, as published. Nobody here has run it
Plan Xiaohongshu Topics
Turn user situations and business evidence into a sustainable publishing system. Do not create a calendar of interchangeable generic topics.
Inputs
Use the account promise, target reader, current account stage, offer, price or decision complexity, recent goal, available proof, material sources, production capacity, existing posts, frequent questions, objections, and desired period. State assumptions when inputs are incomplete.
Topic source map
Build topics from:
- user questions, searches, comparisons, anxieties, and decision criteria;
- work processes, tools, checklists, mistakes, and acceptance rules;
- product use, implementation, limits, service, and aftercare;
- real cases, experiments, before/after mechanisms, and failed assumptions;
- comments, consultations, sales objections, and non-buying reasons;
- policy, market, or product changes that can be verified when current information is required.
Topic jobs
Give every topic one primary job:
| Job | Reader need | Useful formats |
|---|---|---|
| Discovery | “This is about my situation” | scene, tension, mistake, observation |
| Search | explicit question or decision | tutorial, comparison, checklist, FAQ |
| Trust | judge competence and boundaries | process, evidence, case, failure review |
| Proof | validate a claim | demonstration, test, documented result |
| Conversation | contribute an experience or question | judgment, trade-off, prompt |
| Conversion | understand fit and next step | suitability, objection, delivery, FAQ |
| Retention | return for a series | progressive lessons, templates, updates |
Workflow
- Define three to five content pillars from the account promise and proof.
- Translate each pillar into explicit reader situations and search intents.
- Create topics with one job, one reader state, one primary keyword, and one value contract.
- Record the proof or material required. Remove topics that depend on invented evidence.
- Balance the calendar across search answers, execution methods, cases or experiments, and decision checklists.
- Sequence topics so each post creates a useful follow-up rather than repeating the same claim.
- Add a reuse path: comment follow-up, FAQ, carousel, short video, case update, or pinned post.
- Assign one observable learning goal to each post. Do not promise a target result.
Output
Return:
- account promise and content pillars;
- current reader questions and decision stages;
- prioritized topic pool grouped by job;
- calendar with topic, intent, keyword, angle, proof, format, desired action, and learning goal;
- series plan and reuse map;
- material gaps and next production actions.
If the user specifies a publishing period, build that calendar. Otherwise provide a prioritized seven-post sample. When the user requests four posts per week, avoid four versions of one topic: use search answer, execution method, case/experiment, and decision/checklist as the default operating mix unless the evidence suggests a better mix.
Guardrails
Do not invent search volume, trends, cases, user data, platform rules, or performance. Do not treat a fixed posting frequency or content ratio as an algorithm requirement. Do not copy benchmark wording or use unrelated trending terms.
What ships with it: 1 file
255 B alongside SKILL.md
agents/
- openai.yaml255 B