Skills 目录 · 第 77 页

Vite构建工具模式，包括配置、插件、HMR、环境变量、代理设置、SSR、库模式、依赖预打包及构建优化。在处理vite.config.ts、Vite插件或基于Vite的项目时激活。

Vite构建工具模式，包括配置、插件、HMR、环境变量、代理设置、SSR、库模式、依赖预打包及构建优化。在处理vite.config.ts、Vite插件或基于Vite的项目时激活。

用于TypeScript项目（Node、bun、React/Next.js、Effect）中的Vitest：编写、运行或调试单元/组件测试、模拟、测试工具及覆盖率。

语音智能体代表 AI 交互的前沿——人类与 AI 系统自然对话。挑战不仅在于语音识别与合成，还在于在低于 800ms 延迟下实现自然对话流，同时处理打断、背景噪声和情感细微差别。本技能涵盖两种架构：语音到语音（OpenAI Realtime API，延迟最低、最自然）和流水线（STT→LLM→TTS，控制更强、更易调试）。关键洞察：延迟是约束条件。Hu

Expert in building voice AI applications - from real-time voice agents to voice-enabled apps. Covers OpenAI Realtime API, Vapi for voice agents, Deepgram for transcription, ElevenLabs for synthesis, LiveKit for real-time infrastructure, and WebRTC fundamentals. Knows how to build low-latency, production-ready voice experiences. Use when: voice ai, voice agent, speech to text, text to speech, realtime voice.

Write in the Pirate Wires style - authentic, conversational, contrarian. This voice is direct and confident with irreverent humor, takes obvious-but-unsaid positions, and sounds like a smart friend explaining something they've figured out. Use with anti-ai-writing for tech commentary, cultural criticism, or any content requiring a strong point of view.

适用于交付变慢、工程招聘失效、团队结构不清，或需要决定何时增加工程经理的情况。运行 delivery_throughput_analyzer.py（DORA）、eng_hiring_funnel_calculator.py 和 eng_team_structure_designer.py。

就交付速度、工程招聘、团队结构或值班安排对计划提出质疑。在被要求时或使用 /c-level-agents:vpe-review 时使用。

VR/AR 开发原则：舒适度、交互和性能要求。

Build reactive user interfaces with Vue 3: Composition API, reactivity, components, state, and tooling. Use for any Vue-based UI work.

Vue.js 3 Composition API 模式、组件架构、响应式最佳实践、Pinia 状态管理、Vue Router 导航以及 Nuxt SSR 模式。适用于 Vue、Nuxt、Vite 或 Pinia 项目。在构建或审查 Vue 3、Nuxt 或 Pinia 代码时使用——涉及 Composition API、响应式或路由导航。

高级漏洞分析原则。OWASP 2025、供应链安全、攻击面映射、风险优先级排序。

Sweep all recent Claude Code sessions for ones waiting on the user. Use when the user says "sweep my sessions", "what is waiting on me", "which sessions need me", "scan my sessions", or asks to collect open questions across sessions and answer them. Finds sessions that ended on a question, a decision, an approval or a stuck permission prompt, asks the user, drafts each reply for confirmation, then sends it to that session (desktop app) or hands back a resume command (CLI).

Walgreens product data API: search walgreens.com retail products with regular and sale price, unit price, coupons, rewards and rebates, read one product with UPC, ingredients and offers, find Walgreens and Duane Reade stores near a ZIP, and get live unit counts of one product at up to 20 stores in one call. 4 endpoints; search, product and stores 2 credits, stock 4.

Show work as an interactive picture (flow, cloud or Kubernetes map, sequence, state, timeline…) that the user clicks to ask, correct or confirm. Three modes - plan a feature as small steps and build it step by step, explain how existing code works ("explain the login flow in project A"), or compare designs for something new and pick one. Use when the user wants a visual plan, explanation or comparison, wants to follow along before you build, mentions walkboard, or runs /walkboard.

Walmart product data API: keyword search, full product detail by item id, customer reviews with the rating breakdown, category listings, the buy-box offer on an item, marketplace seller storefronts and catalogs, and stores near a US ZIP or Canadian postal code for store-targeted results, as structured JSON. 8 endpoints; 1 credit, or 2 for walmart.com.mx search/category or a store-targeted search/product.

将一大块工作（超出单个智能体会话所能承载的量）规划为问题跟踪器上共享的决策工单地图，并逐一解决，直到通往目标的路径清晰。

Manage Waypoint, the user's shell navigation-and-launch shortcut system (the `waypoint` CLI, tables in ~/.config/waypoint). Use whenever the user wants to add, remove, block, list, or change a project shortcut ('add a waypoint for X', 'make $FOO point to...', 'stop tracking Y', 'what shortcuts do I have'); OR to manage a launch alias that cds into a project and runs a command ('add a launch for X', 'make `launch notes` start claude', 'change the default launch command', 'what can I launch'); when a $VAR shortcut, the `wp` jump, or the `launch` command isn't working in their shell; or when their ~/.zshrc shortcuts broke after an edit. Waypoint renders $VAR exports plus a `wp <name>` cd-jump and a parallel `launch <name>` runner from markdown databases that are the single source of truth.

在提供流动性之前对 DeFi 流动性池进行评分。返回 0-100 的池评分，以及 Enter/Hold/Exit/Reduce/Avoid 结论和经过校准的置信度，覆盖约 68,800 个 Solana 池和 575 个 EVM 池，并提供包含失误的公开历史记录。适用于询问某个池是否值得提供流动性、当前哪些池看起来最好，或这些评分实际准确度如何。

为网站添加 3D 与 WebGL 效果——交互式主视觉对象、产品查看器、粒子、着色器背景、滚动驱动的 3D 场景——可使用 Three.js 或 React Three Fiber（drei）、Spline 或 model-viewer，同时控制加载时间、移动端性能与降级方案。当用户提到 3D、WebGL、Three.js、R3F、react-three-fiber、着色器、GLB/GLTF 模型、粒子、Spline，或想要沉浸式/“惊艳”的主视觉时使用。

审查 UI 代码是否符合 Web 界面指南。当被要求“审查我的 UI”、“检查无障碍性”、“审计设计”、“审查 UX”或“对照最佳实践检查我的网站”时使用。

审查 UI 代码是否符合 Web 界面指南。当被要求“审查我的 UI”、“检查无障碍性”、“审计设计”、“审查 UX”或“对照最佳实践检查我的网站”时使用。

审查 UI 代码是否符合 Web 界面指南。当被要求“审查我的 UI”、“检查无障碍性”、“审计设计”、“审查 UX”或“对照最佳实践检查我的网站”时使用。

This skill enables visual inspection of websites running locally or remotely to identify and fix design issues. Triggers on requests like "review website design", "check the UI", "fix the layout", "find design problems". Detects issues with responsive design, accessibility, visual consistency, and layout breakage, then performs fixes at the source code level.

Web 浏览器游戏开发原则。框架选择、WebGPU、优化、PWA。

诊断并修复网站速度问题——核心网页指标（LCP、INP、CLS）、打包体积、图片、字体、缓存和渲染策略——适用于 Next.js、React/Vite、Vue/Nuxt、Astro 和静态站点。当用户说网站很慢、"website lemot"、想要更好的 Lighthouse/PageSpeed 分数、提到核心网页指标、打包体积过大，或正在为网站上线做准备时使用。

优化网站和 Web 应用性能，包括加载速度、Core Web Vitals、打包体积、缓存策略和运行时性能。

全面的网站质量审计，涵盖性能、可访问性、SEO 和最佳实践。适用于被要求“审计我的网站”、“审查网站质量”、“运行 lighthouse 审计”、“检查页面质量”或“优化我的网站”时。

针对 OWASP Top 10 漏洞的 Web 应用安全测试流程，涵盖注入、XSS、身份认证缺陷和访问控制问题。

Build a simulation-style video (agents, particles, 3D machines, fly-throughs) as one HTML page whose render(t) is a pure function of time, then capture it frame by frame to a 60 fps mp4 with sims/capture.mjs. Use when the idea is a simulation or a 3D object rather than motion graphics with text and UI.

Use ONLY when the user explicitly says: 'use the skill web-to-markdown ...' (or 'use a skill web-to-markdown ...'). Converts webpage URLs to clean Markdown by calling the local web2md CLI (Puppeteer + Readability), suitable for JS-rendered pages.

使用 Playwright 与本地 Web 应用交互并测试的工具包。支持验证前端功能、调试 UI 行为、截取浏览器画面和查看浏览器日志。

在规划、推广或修复网络研讨会或虚拟活动（注册、出席、跟进、转化）时使用。用 webinar_funnel_scorer.py 对每个漏斗阶段评分，找出泄漏环节。

Website crawler API: crawl a whole site to clean Markdown, map every URL from its sitemap, or batch extract up to 1,000 URLs in one job, with a status URL or a signed webhook. Map 1 credit; crawl and batch 1 credit per successful page (2 on ultra), failed pages free; robots.txt obeyed.

Capture a full-page PNG screenshot of any public URL, returned inline as a base64 data:image/png URI ready to drop into an <img> tag. One endpoint, three capture tiers - normal and advanced at 1 credit, ultra for the hardest sites at 5. Only a successful capture is billed. Returns url, image, image_format, mode and image_bytes.

查看网站的外观（桌面和移动端截图）并拆解其背景——颜色、渐变、图像、视频、canvas、::before/::after 伪元素、叠加层和调色板。当用户说“看看这个网站”、“检查一下外观”、“这个网站的背景是什么”、“克隆/复制这个设计”、“从这个网站提取颜色/渐变”，或给出 URL 并询问其视觉效果时使用（也支持印尼语，如“lihat website ini”、“cek tampilan”、“background website apa”、“tiru desain”）。

Build real-time features with WebSockets: connection lifecycle, protocols, backpressure, scaling, and reconnection. Use for live updates and chat.

Turn raw operational data into a weekly management report that answers exactly three questions - what changed, where is it concentrated, what needs a decision. Use when the user mentions weekly report, haftalık rapor, yönetim raporu, ops review, KPI summary, or asks to summarize operational data for managers. Differentiator - contract-first reporting with plain-language findings and driver decomposition, never a chart dump.

适用于有人想执行每周回顾、关闭未闭环事项、审计停滞的项目与承诺、让系统恢复可信状态、重启中断的回顾习惯，或说出“/cs-weekly-review”的情况。按 David Allen 的三阶段循环——GET CLEAR、GET CURRENT、GET CREATIVE——推进，并用确定性脚本盘点未闭环事项、以具名缺口校验清单，并对承诺健康度按 0-100 评分。

Pull Weibo user profiles and posts, post comments/likes/reposts, keyword search across posts, videos, users, topics and images, and the hot-search and ranking boards. 30 endpoints, 1 credit each, structured JSON.

This skill should be used when the user asks "what's behind", "what needs a bump", "what's out of date across our repos", "which products are lagging", "who's on an old version", "portfolio dependency audit", "audit our dependencies across products", "are any products falling behind", "version drift", "which repos have stale actions", "keep the runners updated", or wants a cross-repo currency check spanning EVERY repo in a portfolio rather than one repository. Compares each repo's pinned GitHub Actions, toolchain versions, runner labels, and Dependabot coverage against its peers, then reports which repos are lagging and whether the cause is a missing automation config. Read-only — never edits pins, never opens PRs. For things that are broken or stalled rather than merely old, use whats-on-fire; for single-repo prioritization, use that repository's own survey-work skill.

This skill should be used when the user asks "what's on fire", "what's on fire today", "what's broken", "what's broken across our products", "what's stuck", "what's red", "anything burning", "portfolio status", "portfolio health", "how's the portfolio looking", "which product needs attention", "cross-product status", "any leaked secrets", "security exposure", or wants one cross-product sweep of production failures and stalled work spanning EVERY repo in a GitHub org rather than a single repository. Pulls Sentry issues and cron monitors, org-wide open issues and pull requests, failing workflows, Dependabot exposure, and structural blind spots (products with no error monitoring, no alerting, or no dependency scanning), then ranks them across products and routes each one to the owning repo. Read-only — never files issues, never mutates state. For deep single-repo prioritization, defer to that repository's own survey-work skill.

Set up WhatsApp and read or send messages in the owner's personal account through wa. Use when asked to link or connect WhatsApp, find a WhatsApp chat or person, read a conversation, search messages, or send a message.

Whole Foods grocery data API: search and browse Whole Foods Market at the store you choose with that store's shelf price, sale price and Prime member price with sale dates, list every product on sale at a store, read one product with ingredients, nutrition facts, diets and its in-store and online prices side by side, find stores near a ZIP, read one store, and pull its weekly sales flyer. 7 endpoints, 1 credit each.

将 VitePress/GFM wiki markdown 转换为 Azure DevOps Wiki 兼容格式。生成一个 Node.js 构建脚本，用于转换 Mermaid 语法、去除 front matter、修复链接，并将 ADO-compatible 副本输出到 dist/ado-wiki/。

为仓库文件夹生成 AGENTS.md 文件——包含构建命令、测试说明、代码风格、项目结构和边界的编码智能体上下文文件。仅在 AGENTS.md 缺失时生成。

分析代码仓库并生成带入门指南的分层文档结构。适用于用户想要创建 wiki、生成文档、梳理代码库结构，或从高层理解项目架构时。

分析 git commit 历史并生成按变更类型分类的结构化变更日志。当用户询问近期变更、需要变更日志或想了解仓库中改了什么时使用。

操作任务单要求把已捕获的 `raw/` 来源导入 LLM Wiki，或用户说“ingest <slug>”时使用。创建持久的来源、实体、概念、综合、索引和日志页面；Paperclip 数据包使用 paperclip-distill。

当LLM Wiki操作问题属于lint或健康检查时使用。审计矛盾、孤立页面、来源薄弱、失效链接、缺失的概念页面以及索引/日志漂移；返回分诊发现，不自动修复。