From wewrite
Generates 10 ranked topic ideas for WeChat public accounts by scraping hot topics, analyzing high-frequency demand, and reviewing historical performance. Useful for content planning and avoiding repetitive topics.
How this skill is triggered — by the user, by Claude, or both
Slash command
/wewrite:wewrite-topicThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
- **CLI**:确定性操作走 `wewrite` 命令(需在 PATH;缺失则引导 `uv tool install wewrite`,或在仓库里 `bash install.sh`)。
wewrite 命令(需在 PATH;缺失则引导 uv tool install wewrite,或在仓库里 bash install.sh)。$WEWRITE_HOME 或 ~/.wewrite(wewrite home 可查)。config/style/history/playbook/output/exemplars 全在 {home},不在仓库;references 文档中的状态路径同此约定。读取: <路径> = 用文件读取工具真实读完该文件再继续,不是注释。{skill_dir}/references/;references 文档内的 {skill_dir} 即本 skill 目录。wewrite run show/update/step 读取和更新;不直接改状态文件。{home}/style.yaml 存在 → 提取 topics、content_style。不存在 → 问用户"你的
公众号主要写哪几个方向?"用回答临时代替 topics,并提示可用 wewrite-style 完成完整设置。wewrite hotspots --limit 30
降级:脚本报错 → WebSearch "今日热点 {topics第一个垂类}"
对 topics 的前 1-2 个垂类词各搜一次近一周的公众号文章:
wewrite search-articles "{垂类词}" -n 15 -t 2
产出近一周同垂类公众号文章列表(标题/摘要/发布时间/账号),作为「已验证的内容需求」 信号供 2.3 使用。搜狗结果不带阅读量,因此只能说明同题出现频率,不能声称文章“爆款”。
降级:命令报错 / 结果为空(搜狗反爬限流)→ 跳过本步,不阻断,选题构成退回默认。
读取: {home}/history.yaml(不存在则跳过)
wewrite seo --json {关键词}
历史分析(有 stats 数据时):
framework 的文章表现最好(阅读量/分享率)→ 推荐框架时加权enhance_strategy 的文章表现最好 → 增强策略选择时参考降级:SEO 脚本报错 → LLM 判断;history 无 stats → 跳过效果分析,仅做去重
读取: {skill_dir}/references/topic-selection.md
生成 10 个选题,其中:
topics 领域生成长尾内容(教程/方法论/经验总结/工具推荐),标注为"常青"。适合 content_style 为干货型/测评型的用户每个选题含标题、评分、点击率潜力、SEO 友好度、推荐框架。
用 wewrite run update --patch 写入 topic.title、topic.keywords、topic.source
(“热点抓取” / “高频需求” / “常青”)、topic.framework_hint,然后
wewrite run step topic completed。
单独激活且用户只要选题列表时,展示 10 个选题即可;用户选定后写入状态并提示
"可以直接说'就写这个'进入写作(wewrite-write)"。
npx claudepluginhub imraywang/wewriteAutomates end-to-end WeChat Official Account article creation: fetches hotspots, selects topics, writes, optimizes SEO, generates images, formats, and publishes to draft box.
Plans data-driven content strategies by analyzing search demand, community questions, and shareable formats via UnifAPI. Helps decide what to write about based on public demand.
Automates WeChat public account article writing: search relevant materials, compose 1000-1500 word articles, generate viral headlines, and optimize formatting. Useful when creating WeChat articles or self-media content.