每日工作回顾与洞察分析，对用户前一天的对话历史和使用数据进行总结与建议
npx skills add https://github.com/zhangdszq/vk-skills --skill daily-review复制此命令并粘贴到 Claude Code 中以安装该技能
Use this skill whenever the user wants sales-centric analysis driven by the `intl_analysis` StarRocks database, especially from a Saudi `leads-to-order` conversion perspective, and enriched by Megaview conversation evidence. Use it for seller performance comparison, session score and customer score analysis, conversation count analysis, GMV comparison, training samples, coaching calls, transcript snippets, manager coaching material, or to identify who should enter performance review first. Use it when the user wants to start from the growth chain in `intl_analysis` - channel, marketing, sales team, LP, trial completion, post-trial follow-up, TCC, and order conversion - then use Megaview to explain seller quality, fetch representative conversations, or build a review-priority list for underperformers. This skill is specifically for sales and growth-chain analysis built on `intl_analysis` plus Megaview, not for general-purpose Megaview API documentation, request drafting, or ad-hoc endpoint execution.
Extract teaching content from courseware files by following a user-provided dimension table or rubric, especially when the user gives an Excel/worksheet with categories and extraction rules plus a PPT/PPTX to analyze. Use this whenever the user wants you to map slide content into dimensions such as vocabulary, reading, phonics, grammar, sentence structures, or similar curriculum buckets, even if they do not explicitly ask for a "skill." Also use it when the user wants the extracted result filled into a new Excel workbook, wants unfound items left blank, wants a structured teaching-content summary generated from slide decks, or asks to use the built-in L4 teaching-dimension standard without re-uploading the Excel. When the input includes a PPT/PPTX, first use the PPTX skill workflow to turn the deck into per-slide images, then extract from both slide images and text together rather than relying on text alone.
Best practices for Remotion - Video creation in React
VIPKID 国际化运营后台助理。商品包等国际化管理能力操作 sa-manager.lionabc.com； Leads 管理页 `/leadsGCC/list`、私池/公海/冻结库列表查询与相关接口操作 sa-leads.lionabc.com。 支持商品包查询、新建、修改、库存、优惠券限制、赠送权限； 也支持 omnicenter 家长详情页相关操作，包括家长详情、资料修改、备注联系方式、时区、推荐码、解密、课耗、学习行为、跟进记录、海报配置查询； 还支持 Leads 列表筛选查询、节点统计、标签管理、解密用户、批量分配/流转、上传 Leads/权益、家长详情与关联页面入口说明。 用户说「查商品包」「新建课包」「修改价格」「配置库存」「优惠券限制」「赠送配置」「家长详情」「修改家长信息」「查课耗」「查学习行为」「查跟进记录」「查 leads」「Leads管理」「分配 GCC/GCS/TMK」「冻结 leads」「流转公海」「跟进 leads」「线索库」「解密用户」「上传 leads」时触发。 如果用户是口语化提问，比如「公海前10」「今天注册多少」「待付费从 2.1 到现在」「这个脚本在哪」「skill 该怎么改」，也要主动触发并优先用现成脚本回答，不要只给页面说明。 ⛔ 绝不执行上架、下架、删除操作；任何批量分配、冻结、流转、上传、改密前必须二次确认。
Converts Markdown to styled HTML with WeChat-compatible themes. Supports code highlighting, math, PlantUML, footnotes, alerts, and infographics. Use when user asks for "markdown to html", "convert md to html", "md转html", or needs styled HTML output from markdown.
Posts content to WeChat Official Account (微信公众号) via API or Chrome CDP. Supports article posting (文章) with HTML, markdown, or plain text input, and image-text posting (贴图, formerly 图文) with multiple images. Use when user mentions "发布公众号", "post to wechat", "微信公众号", or "贴图/图文/文章".
| name | daily-review |
| description | 每日工作回顾与洞察分析，对用户前一天的对话历史和使用数据进行总结与建议 |
你是一个 AI 工作回顾助手。用户会给你一份结构化的昨日工作数据（JSON 格式），请按以下要求生成回顾报告。
JSON 数据包含以下字段：
date: 分析日期workspaces: 工作区数组，每个包含：
workspaceName: 工作区/项目名称conversations: 压缩后的对话列表（含标题、用户问题、AI 摘要、工具使用、未完成标记）usage: Token 消耗和模型使用统计globalStats: 全局统计（总对话数、时间分配）对每个有活动的工作区，输出以下内容（根据 modules 字段决定包含哪些）：
lastMessageRole 为 user 表示对话可能中断containsTodo 为 true 表示对话中提到了待办事项