From wewrite
Pulls reading/sharing/like stats from WeChat Official Account articles, backfills history, and suggests topic, title, and framework adjustments.
How this skill is triggered — by the user, by Claude, or both
Slash command
/wewrite:wewrite-statsThis 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。读取: {skill_dir}/references/effect-review.md
按其流程执行:fetch_stats.py --days 7 拉数据 → 匹配并回填 history.yaml 的
stats 字段 → 分析最好/最差表现及原因 → 给出后续选题/标题/框架的调整建议。
前置:需要 config.yaml 里的微信 API 凭证。缺凭证 → 告知用户"数据复盘需要配置 公众号 API(config.yaml),当前只能基于 history.yaml 已有记录做定性分析",然后就 history.yaml 现有内容能分析多少分析多少。刚发布的文章 → 告知等 24h 后再看。
下游影响:回填的 stats 会被选题模块(wewrite-topic)读取——哪种框架/增强策略 表现好会加权到下次推荐。这是数据闭环的一半,另一半是 wewrite-learn 的改稿飞轮。
npx claudepluginhub imraywang/wewriteGenerates 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.
Automates end-to-end WeChat Official Account article creation: fetches hotspots, selects topics, writes, optimizes SEO, generates images, formats, and publishes to draft box.
Provides structured methods for content operations: topic competition analysis, title scoring/optimization, multi-platform content adaptation, and weekly content review. Useful for content creators and marketers.