1961 lines
230 KiB
Markdown
1961 lines
230 KiB
Markdown
<!-- WEHUB_ZH_README -->
|
||
> [!NOTE]
|
||
> 本文档由 WeHub 基于上游 README 翻译整理,属于社区翻译,非官方中文文档。
|
||
> [English](./README.en.md) · [原始项目](https://github.com/VoltAgent/awesome-agent-skills) · [上游 README](https://github.com/VoltAgent/awesome-agent-skills/blob/HEAD/README.md)
|
||
> 原作者、版权与许可证归属以原始项目及本仓库 LICENSE 文件为准。
|
||
|
||
<a href="https://github.com/VoltAgent/voltagent">
|
||
<img width="1500" alt="claude-skills" src="https://github.com/user-attachments/assets/0db54cfc-f3dd-4683-abbb-e4c01d9dfb5d" />
|
||
</a>
|
||
|
||
|
||
<br/>
|
||
<br/>
|
||
|
||
<div align="center">
|
||
<strong>来自领先开发团队与社区的官方 Agent Skills 合集。
|
||
<br />
|
||
人工精选,而非 AI 批量生成。
|
||
</strong>
|
||
<br />
|
||
<br />
|
||
|
||
</div>
|
||
|
||
<div align="center">
|
||
|
||
[](https://awesome.re)
|
||

|
||

|
||
<a href="https://github.com/VoltAgent/voltagent">
|
||
<img alt="VoltAgent" src="https://cdn.voltagent.dev/website/logo/logo-2-svg.svg" height="20" />
|
||
</a>
|
||
[](https://s.voltagent.dev/discord)
|
||
|
||
|
||
</div>
|
||
|
||
</div>
|
||
|
||
# Awesome Agent Skills
|
||
|
||
与许多批量生成的技能仓库不同,本合集聚焦于真实工程团队创建并使用的实战 Agent Skills,而非 AI 大规模生成的内容。
|
||
|
||
本合集收录了领先开发团队发布的官方技能,包括 Anthropic、Google Labs、Vercel、Stripe、Cloudflare、Netlify、Trail of Bits、Sentry、Expo、Hugging Face、Figma 等,以及社区构建的技能。
|
||
|
||
兼容 Claude Code、Codex、Antigravity、Gemini CLI、Cursor、GitHub Copilot、OpenCode、Windsurf 等。路径与文档请见下表。
|
||
|
||
贡献最多的 Agent Skills 仓库,与社区共同构建与维护。
|
||
|
||
|
||
## 💛 赞助商
|
||
|
||
| | |
|
||
| :-: | :-- |
|
||
| <a href="https://www.testmuai.com"><picture><source media="(prefers-color-scheme: dark)" srcset="https://cdn.voltagent.dev/awesome-repo/testmui/testmuai-white.png"><img alt="TestMu AI" src="https://cdn.voltagent.dev/awesome-repo/testmui/testmuai-black.png" width="425"></picture></a> | [TestMu AI(前身为 LambdaTest)](https://www.testmuai.com) 是一款面向现代工程团队的 AI 原生测试云平台,涵盖从自主测试创建与快速执行,到测试 AI 智能体、聊天机器人和语音助手等全流程。 |
|
||
| <a href="https://zero.xyz"><picture><source media="(prefers-color-scheme: dark)" srcset="https://cdn.voltagent.dev/awesome-repo/zero-xyz/zero-dark.png"><img alt="Zero" src="https://cdn.voltagent.dev/awesome-repo/zero-xyz/zero-light.png" width="425"></picture></a> | [Zero](https://zero.xyz) 让你的 AI 访问数千种工具、API 和服务,无需配置即可从提示词直达项目。无需让你注册或获取 API 密钥,你的 AI 可直接在聊天中发现并使用能解决实际问题的真实服务。 |
|
||
| <a href="https://lite.ego.app/?utm_source=officialskills&utm_medium=sponsor&utm_campaign=github-sponsor"><picture><source media="(prefers-color-scheme: dark)" srcset="https://cdn.voltagent.dev/awesome-repo/ego-lite/logo_wordmark_lite_white.svg"><img alt="Ego Lite" src="https://cdn.voltagent.dev/awesome-repo/ego-lite/logo_wordmark_lite_dark.svg" width="425"></picture></a> | [Ego Lite](https://lite.ego.app/?utm_source=officialskills&utm_medium=sponsor&utm_campaign=github-sponsor) 是 AI 智能体运行浏览器自动化任务的最快浏览器,比 agent-browser(Vercel)快 3.45 倍,永久免费、无需配置,并支持智能体在 Spaces 中同时运行 100+ 个浏览器任务。 |
|
||
|
||
<br />
|
||
|
||
<a href="https://sponsors.voltagent.dev/#awesome-agent-skills"><img src="https://img.shields.io/badge/📩_Become_a_Sponsor-Contact_Us-blue?style=for-the-badge&logoColor=white" alt="Become a Sponsor" /></a>
|
||
|
||
|
||
## 目录
|
||
|
||
### 各团队官方技能
|
||
|
||
| | | | |
|
||
|---|---|---|---|
|
||
| [Claude](#official-claude-skills) | [VoltAgent](#skills-by-voltagent) | [TestMu AI](#skills-by-testmu-ai) | [Zero](#skills-by-zero) |
|
||
| [Angular](#skills-by-angular) | [Composio](#skills-by-composio-team) | [Supabase](#skills-by-supabase-team) | [Google Gemini](#skills-by-google-gemini) |
|
||
| [Stripe](#skills-by-stripe-team) | [Courier](#skills-by-courier) | [CallStack](#skills-by-callstack) | [Expo](#skills-by-expo-team) |
|
||
| [Better Auth](#skills-by-better-auth-team) | [Tinybird](#skills-by-tinybird-team) | [HashiCorp](#skills-by-hashicorp-team-for-terraform) | [Sanity](#skills-by-sanity-team) |
|
||
| [Firecrawl](#skills-by-firecrawl-team) | [Neon](#skills-by-neon-team) | [ClickHouse](#skill-by-clickhouse) | [Remotion](#skills-by-remotion) |
|
||
| [Replicate](#skills-by-replicate) | [Typefully](#skills-by-typefully) | [Vercel](#skills-by-vercel-engineering-team) | [Cloudflare](#skills-by-cloudflare-team) |
|
||
| [Netlify](#skills-by-netlify-team) | [Google Labs (Stitch)](#skills-by-google-labs-stitch) | [Google Workspace CLI](#skills-by-google-workspace-cli) | [Hugging Face](#skills-by-hugging-face-team) |
|
||
| [Trail of Bits](#security-skills-by-trail-of-bits-team) | [Sentry](#skills-by-sentry-team-for-their-dev-team) | [Microsoft](#skills-by-microsoft) | [fal.ai](#skills-by-falai-team) |
|
||
| [WordPress](#skills-by-wordpress-development-team) | [OpenAI](#skills-by-openai) | [Figma](#skills-by-figma) | [Corey Haines](#marketing-skills-by-corey-haines) |
|
||
| [Binance](#skills-by-binance) | [Dean Peters](#product-manager-skills-by-dean-peters) | [Paweł Huryn](#product-management-skills-by-pawel-huryn) | [MiniMax](#skills-by-minimax-team) |
|
||
| [DuckDB](#skills-by-duckdb) | [GSAP](#skills-by-gsap-greensock) | [Garry Tan (gstack)](#skills-by-garry-tan-gstack) | [Notion](#skills-by-notion) |
|
||
| [Resend](#skills-by-resend) | [Addy Osmani (Web Quality)](#skills-by-addy-osmani-web-quality) | [MongoDB](#skills-by-mongodb) | [Kim Barrett (Advertising)](#advertising-skills-by-kim-barrett) |
|
||
| [Apollo GraphQL](#skills-by-apollo-graphql) | [Auth0](#skills-by-auth0) | [Brave](#skills-by-brave) | [Browserbase](#skills-by-browserbase) |
|
||
| [CodeRabbit](#skills-by-coderabbit) | [Coinbase](#skills-by-coinbase) | [Datadog Labs](#skills-by-datadog-labs) | [Firebase](#skills-by-firebase) |
|
||
| [Flutter](#skills-by-flutter) | [Venice.ai](#skills-by-veniceai) | [Red Hat](#skills-by-redhat) | [Community](#community-skills) |
|
||
| [Redis](#skills-by-redis) | [NVIDIA](#skills-by-nvidia) | [Google Cloud](#skills-by-google-cloud) | [Quality Standards](#skill-quality-standards) |
|
||
|
||
|
||
<br/>
|
||
|
||
<br/>
|
||
|
||
<a href="https://launchkit.getdesign.md/">
|
||
<img src="https://cdn.voltagent.dev/awesome-repo/launchkit-banner-3.png" alt="launchkit" /><br/>
|
||
</a>
|
||
|
||
<br/>
|
||
|
||
|
||
<br/>
|
||
|
||
<details open>
|
||
<summary><h3 style="display:inline">官方 Claude 技能</h3></summary>
|
||
|
||
- **[anthropics/docx](https://officialskills.sh/anthropics/skills/docx)** - 创建、编辑和分析 Word 文档
|
||
- **[anthropics/doc-coauthoring](https://officialskills.sh/anthropics/skills/doc-coauthoring)** - 协作文档编辑与共同撰写
|
||
- **[anthropics/pptx](https://officialskills.sh/anthropics/skills/pptx)** - 创建、编辑和分析 PowerPoint 演示文稿
|
||
- **[anthropics/xlsx](https://officialskills.sh/anthropics/skills/xlsx)** - 创建、编辑和分析 Excel 电子表格
|
||
- **[anthropics/pdf](https://officialskills.sh/anthropics/skills/pdf)** - 提取文本、创建 PDF 并处理表单
|
||
- **[anthropics/algorithmic-art](https://officialskills.sh/anthropics/skills/algorithmic-art)** - 使用 p5.js 与种子随机数创建生成艺术
|
||
- **[anthropics/canvas-design](https://officialskills.sh/anthropics/skills/canvas-design)** - 以 PNG 和 PDF 格式设计视觉艺术
|
||
- **[anthropics/frontend-design](https://officialskills.sh/anthropics/skills/frontend-design)** - 前端设计与 UI/UX 开发工具
|
||
- **[anthropics/slack-gif-creator](https://officialskills.sh/anthropics/skills/slack-gif-creator)** - 创建针对 Slack 尺寸限制优化的动画 GIF
|
||
- **[anthropics/theme-factory](https://officialskills.sh/anthropics/skills/theme-factory)** - 使用专业主题为制品设置样式或生成自定义主题
|
||
- **[anthropics/web-artifacts-builder](https://officialskills.sh/anthropics/skills/web-artifacts-builder)** - 使用 React 和 Tailwind 构建复杂的 claude.ai HTML 制品
|
||
- **[anthropics/mcp-builder](https://officialskills.sh/anthropics/skills/mcp-builder)** - 创建 MCP 服务器以集成外部 API 和服务
|
||
- **[anthropics/webapp-testing](https://officialskills.sh/anthropics/skills/webapp-testing)** - 使用 Playwright 测试本地 Web 应用
|
||
- **[anthropics/brand-guidelines](https://officialskills.sh/anthropics/skills/brand-guidelines)** - 将 Anthropic 品牌色彩与字体应用于制品
|
||
- **[anthropics/internal-comms](https://officialskills.sh/anthropics/skills/internal-comms)** - 撰写状态报告、新闻通讯和常见问题解答
|
||
- **[anthropics/skill-creator](https://officialskills.sh/anthropics/skills/skill-creator)** - 创建扩展 Claude 能力的技能指南
|
||
- **[anthropics/template](https://officialskills.sh/anthropics/skills/template)** - 创建新技能的基础模板
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">VoltAgent 技能</h3></summary>
|
||
|
||
VoltAgent 官方技能,用于基于 VoltAgent TypeScript 框架构建 AI agent。
|
||
- **[voltagent/create-voltagent](https://officialskills.sh/voltagent/skills/create-voltagent)** - 项目设置指南,包含 CLI 与手动步骤
|
||
- **[voltagent/voltagent-best-practices](https://officialskills.sh/voltagent/skills/voltagent-best-practices)** - 面向 agent、工作流、记忆与服务器的架构与使用模式
|
||
- **[voltagent/voltagent-core-reference](https://officialskills.sh/voltagent/skills/voltagent-core-reference)** - VoltAgent 类选项与生命周期方法参考
|
||
- **[voltagent/voltagent-docs-bundle](https://officialskills.sh/voltagent/skills/voltagent-docs-bundle)** - 从 @voltagent/core 查询嵌入式文档,获取与版本匹配的文档
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">TestMu AI 技能</h3></summary>
|
||
|
||
面向各主流测试自动化框架的生产级 Agent Skills,由 TestMu AI(前身为 LambdaTest)团队维护。它们可帮助 AI 编码助手在 Web、移动端、API、BDD 与单元测试技术栈中生成专家级测试自动化代码。
|
||
|
||
- **[testmu-ai/api-skill](https://github.com/LambdaTest/agent-skills/tree/main/api-skill)** - 一套 API 技能,用于设计、模拟、文档化、保护 REST/GraphQL/gRPC API,并生成相关测试
|
||
- **[testmu-ai/appium-skill](https://github.com/LambdaTest/agent-skills/tree/main/appium-skill)** - 使用 Java、Python 或 JS 生成面向 Android 与 iOS 的 Appium 移动端自动化测试
|
||
- **[testmu-ai/behat-skill](https://github.com/LambdaTest/agent-skills/tree/main/behat-skill)** - 使用 Gherkin 与 Mink 为 PHP 生成 Behat BDD 测试
|
||
- **[testmu-ai/behave-skill](https://github.com/LambdaTest/agent-skills/tree/main/behave-skill)** - 使用 Gherkin 与步骤实现为 Python 生成 Behave BDD 测试
|
||
- **[testmu-ai/capybara-skill](https://github.com/LambdaTest/agent-skills/tree/main/capybara-skill)** - 在 Ruby 中生成集成 RSpec 的 Capybara E2E 测试
|
||
- **[testmu-ai/cicd-pipeline-skill](https://github.com/LambdaTest/agent-skills/tree/main/cicd-pipeline-skill)** - 在 GitHub Actions、Jenkins、GitLab CI 与 Azure DevOps 上为测试生成 CI/CD 流水线
|
||
- **[testmu-ai/codeception-skill](https://github.com/LambdaTest/agent-skills/tree/main/codeception-skill)** - 在 PHP 中生成 Codeception 验收、功能与单元测试
|
||
- **[testmu-ai/cucumber-skill](https://github.com/LambdaTest/agent-skills/tree/main/cucumber-skill)** - 使用 Gherkin 与步骤定义在 Java、JS 或 Ruby 中生成 Cucumber BDD 测试
|
||
- **[testmu-ai/cypress-skill](https://github.com/LambdaTest/agent-skills/tree/main/cypress-skill)** - 使用 JavaScript 或 TypeScript 生成 Cypress E2E 与组件测试
|
||
- **[testmu-ai/detox-skill](https://github.com/LambdaTest/agent-skills/tree/main/detox-skill)** - 使用 JavaScript 为 React Native 应用生成 Detox 灰盒 E2E 测试
|
||
- **[testmu-ai/espresso-skill](https://github.com/LambdaTest/agent-skills/tree/main/espresso-skill)** - 使用 Kotlin 或 Java 为 Android 应用生成 Espresso UI 测试
|
||
- **[testmu-ai/flutter-testing-skill](https://github.com/LambdaTest/agent-skills/tree/main/flutter-testing-skill)** - 使用 Dart 生成 Flutter widget、集成与 golden 测试
|
||
- **[testmu-ai/gauge-skill](https://github.com/LambdaTest/agent-skills/tree/main/gauge-skill)** - 生成 Markdown 格式的 Gauge 规范,步骤可使用 Java、Python、JS 或 Ruby
|
||
- **[testmu-ai/geb-skill](https://github.com/LambdaTest/agent-skills/tree/main/geb-skill)** - 使用 Groovy 配合 Spock 与页面对象生成 Geb 浏览器自动化测试
|
||
- **[testmu-ai/hyperexecute-skill](https://github.com/LambdaTest/agent-skills/tree/main/hyperexecute-skill)** - 端到端操作 TestMu AI HyperExecute:YAML、CLI 运行、调试与 CI 集成
|
||
- **[testmu-ai/jasmine-skill](https://github.com/LambdaTest/agent-skills/tree/main/jasmine-skill)** - 使用 JavaScript 生成支持 spies 与异步的 Jasmine BDD 测试
|
||
- **[testmu-ai/jest-skill](https://github.com/LambdaTest/agent-skills/tree/main/jest-skill)** - 在 JS/TS 中生成支持 mock 与快照的 Jest 单元与集成测试
|
||
- **[testmu-ai/junit-5-skill](https://github.com/LambdaTest/agent-skills/tree/main/junit-5-skill)** - 使用 Java 配合 Mockito 生成 JUnit 5 单元与集成测试
|
||
- **[testmu-ai/karma-skill](https://github.com/LambdaTest/agent-skills/tree/main/karma-skill)** - 为基于浏览器的 JS 测试生成 Karma 测试运行器配置
|
||
- **[testmu-ai/laravel-dusk-skill](https://github.com/LambdaTest/agent-skills/tree/main/laravel-dusk-skill)** - 在 PHP 中生成基于 Chrome 的 Laravel Dusk 浏览器测试
|
||
- **[testmu-ai/lettuce-skill](https://github.com/LambdaTest/agent-skills/tree/main/lettuce-skill)** - 为 Python 生成 Lettuce BDD 测试(旧版;建议优先使用 Behave)
|
||
- **[testmu-ai/mocha-skill](https://github.com/LambdaTest/agent-skills/tree/main/mocha-skill)** - 使用 JavaScript 配合 Chai 与 Sinon 生成 Mocha 测试
|
||
- **[testmu-ai/mstest-skill](https://github.com/LambdaTest/agent-skills/tree/main/mstest-skill)** - 为 .NET 使用 C# 生成 MSTest 测试
|
||
- **[testmu-ai/nemojs-skill](https://github.com/LambdaTest/agent-skills/tree/main/nemojs-skill)** - 为 Node.js 生成基于 Selenium 的 Nemo.js 测试
|
||
- **[testmu-ai/nightwatchjs-skill](https://github.com/LambdaTest/agent-skills/tree/main/nightwatchjs-skill)** - 使用 JavaScript 配合 Selenium WebDriver 生成 NightwatchJS E2E 测试
|
||
- **[testmu-ai/nunit-skill](https://github.com/LambdaTest/agent-skills/tree/main/nunit-skill)** - 使用 C# 配合约束模型与 Moq 生成 NUnit 3 测试
|
||
- **[testmu-ai/phpunit-skill](https://github.com/LambdaTest/agent-skills/tree/main/phpunit-skill)** - 在 PHP 中生成支持数据提供器与 mock 的 PHPUnit 测试
|
||
- **[testmu-ai/playwright-skill](https://github.com/LambdaTest/agent-skills/tree/main/playwright-skill)** - 使用 TS、JS、Python、Java 或 C# 生成 Playwright E2E 测试
|
||
- **[testmu-ai/protractor-skill](https://github.com/LambdaTest/agent-skills/tree/main/protractor-skill)** - 使用 JS/TS 为 Angular 生成 Protractor E2E 测试(已弃用;建议优先使用 Playwright/Cypress)
|
||
- **[testmu-ai/puppeteer-skill](https://github.com/LambdaTest/agent-skills/tree/main/puppeteer-skill)** - 生成用于浏览器自动化、爬取与 PDF 生成的 Puppeteer 脚本
|
||
- **[testmu-ai/pytest-skill](https://github.com/LambdaTest/agent-skills/tree/main/pytest-skill)** - 在 Python 中生成支持 fixtures、parametrize 与 mock 的 pytest 测试
|
||
- **[testmu-ai/reqnroll-skill](https://github.com/LambdaTest/agent-skills/tree/main/reqnroll-skill)** - 使用 C# 为 Web 与移动端生成 Reqnroll BDD 测试
|
||
- **[testmu-ai/robot-framework-skill](https://github.com/LambdaTest/agent-skills/tree/main/robot-framework-skill)** - 在 Python 中生成 Robot Framework 关键字驱动测试
|
||
- **[testmu-ai/rspec-skill](https://github.com/LambdaTest/agent-skills/tree/main/rspec-skill)** - 在 Ruby 中生成支持 matchers、hooks 与 mock 的 RSpec 测试
|
||
- **[testmu-ai/selenide-skill](https://github.com/LambdaTest/agent-skills/tree/main/selenide-skill)** - 在 Java 中生成支持自动等待与流式 API 的 Selenide UI 测试
|
||
- **[testmu-ai/selenium-skill](https://github.com/LambdaTest/agent-skills/tree/main/selenium-skill)** - 使用 Java、Python、JS、C#、Ruby 或 PHP 生成 Selenium WebDriver 测试
|
||
- **[testmu-ai/serenity-bdd-skill](https://github.com/LambdaTest/agent-skills/tree/main/serenity-bdd-skill)** - 在 Java 中使用 Screenplay 模式与报告能力生成 Serenity BDD 测试
|
||
- **[testmu-ai/smartui-skill](https://github.com/LambdaTest/agent-skills/tree/main/smartui-skill)** - 生成用于截图比对的 SmartUI 视觉回归配置
|
||
- **[testmu-ai/specflow-skill](https://github.com/LambdaTest/agent-skills/tree/main/specflow-skill)** - 使用 Gherkin 与步骤绑定为 C#/.NET 生成 SpecFlow BDD 测试
|
||
- **[testmu-ai/test-framework-migration-skill](https://github.com/LambdaTest/agent-skills/tree/main/test-framework-migration-skill)** - 在 Selenium、Playwright、Puppeteer 与 Cypress 之间迁移测试
|
||
- **[testmu-ai/testcafe-skill](https://github.com/LambdaTest/agent-skills/tree/main/testcafe-skill)** - 使用 JavaScript 或 TypeScript 生成 TestCafe 自动化测试
|
||
- **[testmu-ai/testng-skill](https://github.com/LambdaTest/agent-skills/tree/main/testng-skill)** - 在 Java 中生成支持数据提供器与并行执行的 TestNG 测试
|
||
- **[testmu-ai/testunit-skill](https://github.com/LambdaTest/agent-skills/tree/main/testunit-skill)** - 在 Ruby 中生成 Test::Unit xUnit 风格测试
|
||
- **[testmu-ai/unittest-skill](https://github.com/LambdaTest/agent-skills/tree/main/unittest-skill)** - 生成使用 TestCase 与 setUp/tearDown 的 Python unittest 测试
|
||
- **[testmu-ai/vitest-skill](https://github.com/LambdaTest/agent-skills/tree/main/vitest-skill)** - 在 JS/TS 中生成具备 Jest 兼容 API 与 ESM 支持的 Vitest 测试
|
||
- **[testmu-ai/webdriverio-skill](https://github.com/LambdaTest/agent-skills/tree/main/webdriverio-skill)** - 使用 JavaScript 或 TypeScript 生成 WebdriverIO(WDIO)自动化测试
|
||
- **[testmu-ai/xcuitest-skill](https://github.com/LambdaTest/agent-skills/tree/main/xcuitest-skill)** - 使用 Swift 为 iOS/iPadOS 应用生成 XCUITest UI 测试
|
||
- **[testmu-ai/xunit-skill](https://github.com/LambdaTest/agent-skills/tree/main/xunit-skill)** - 使用 C# 配合 Fact/Theory 与 FluentAssertions 生成 xUnit.net 测试
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Zero 的技能</h3></summary>
|
||
|
||
|
||
- **[zero/zero](https://github.com/officialzeroxyz/zero-plugins/blob/main/plugins/zero/skills/zero/SKILL.md)** - 为 Claude Code 代理发现并调用外部付费工具,无需中断流程让用户注册或获取 API 密钥
|
||
- **[zero/zero-gemini](https://github.com/officialzeroxyz/zero-plugins/tree/main/plugins/zero-gemini)** - 将相同的 Zero 工具发现与支付层打包为 Gemini CLI 扩展
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Angular 的技能</h3></summary>
|
||
|
||
- **[angular/angular-developer](https://github.com/angular/skills)** - 为 components、services、reactivity 生成 Angular 代码与架构指导
|
||
- **[angular/angular-new-app](https://github.com/angular/skills)** - 使用 CLI 按现代最佳实践创建新的 Angular 应用
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Composio 团队的技能</h3></summary>
|
||
|
||
- **[composiohq/composio](https://officialskills.sh/composiohq/skills/composio)** - 通过托管认证将 AI 代理连接至 1000+ 外部应用
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Supabase 团队的技能</h3></summary>
|
||
|
||
- **[supabase/postgres-best-practices](https://officialskills.sh/supabase/skills/postgres-best-practices)** - Supabase 的 PostgreSQL 最佳实践
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Google Gemini 的技能</h3></summary>
|
||
|
||
- **[google-gemini/gemini-api-dev](https://officialskills.sh/google-gemini/skills/gemini-api-dev)** - 使用 Gemini API 开发 Gemini 驱动应用的最佳实践
|
||
- **[google-gemini/vertex-ai-api-dev](https://officialskills.sh/google-gemini/skills/vertex-ai-api-dev)** - 使用 Gen AI SDK 在 Google Cloud Vertex AI 上开发 Gemini 驱动应用
|
||
- **[google-gemini/gemini-live-api-dev](https://officialskills.sh/google-gemini/skills/gemini-live-api-dev)** - 使用 Gemini Live API 构建实时双向流式应用
|
||
- **[google-gemini/gemini-interactions-api](https://officialskills.sh/google-gemini/skills/gemini-interactions-api)** - 使用 Gemini Interactions API 构建支持文本、聊天、流式传输和图像生成的应用
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Stripe 团队的技能</h3></summary>
|
||
|
||
- **[stripe/stripe-best-practices](https://officialskills.sh/stripe/skills/stripe-best-practices)** - 构建 Stripe 集成的最佳实践
|
||
- **[stripe/upgrade-stripe](https://officialskills.sh/stripe/skills/upgrade-stripe)** - 升级 Stripe SDK 与 API 版本
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Courier 的技能</h3></summary>
|
||
|
||
- **[trycourier/courier-skills](https://github.com/trycourier/courier-skills)** - 通过电子邮件、SMS、推送和聊天实现多渠道通知
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">CallStack 的技能</h3></summary>
|
||
|
||
- **[callstackincubator/react-native-best-practices](https://officialskills.sh/callstackincubator/skills/react-native-best-practices)** - 来自 Callstack 的 React Native 应用性能优化
|
||
- **[callstackincubator/github](https://officialskills.sh/callstackincubator/skills/github)** - 用于 PR、代码评审与分支的 GitHub 工作流模式
|
||
- **[callstackincubator/upgrading-react-native](https://officialskills.sh/callstackincubator/skills/upgrading-react-native)** - React Native 升级工作流:模板、依赖项与常见陷阱
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Better Auth 团队的技能</h3></summary>
|
||
|
||
- **[better-auth/best-practices](https://officialskills.sh/better-auth/skills/best-practices)** - Better Auth 集成的最佳实践
|
||
- **[better-auth/explain-error](https://officialskills.sh/better-auth/skills/explain-error)** - 解释 Better Auth 错误消息
|
||
- **[better-auth/providers](https://officialskills.sh/better-auth/skills/providers)** - Better Auth 认证提供方
|
||
- **[better-auth/create-auth](https://officialskills.sh/better-auth/skills/create-auth)** - 使用 Better Auth 创建认证配置
|
||
- **[better-auth/emailAndPassword](https://officialskills.sh/better-auth/skills/emailAndPassword)** - 使用 Better Auth 实现电子邮件与密码认证
|
||
- **[better-auth/organization](https://officialskills.sh/better-auth/skills/organization)** - 使用 Better Auth 进行组织管理
|
||
- **[better-auth/twoFactor](https://officialskills.sh/better-auth/skills/twoFactor)** - 使用 Better Auth 实现双因素认证
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Tinybird 团队的技能</h3></summary>
|
||
|
||
- **[tinybirdco/tinybird-best-practices](https://officialskills.sh/tinybirdco/skills/tinybird-best-practices)** - Tinybird 项目中 datasources、pipes、endpoints 与 SQL 的指南
|
||
- **[tinybirdco/tinybird-cli-guidelines](https://officialskills.sh/tinybirdco/skills/tinybird-cli-guidelines)** - Tinybird CLI 使用指南与命令
|
||
- **[tinybirdco/tinybird-python-sdk-guidelines](https://officialskills.sh/tinybirdco/skills/tinybird-python-sdk-guidelines)** - Tinybird Python SDK 使用指南
|
||
- **[tinybirdco/tinybird-typescript-sdk-guidelines](https://officialskills.sh/tinybirdco/skills/tinybird-typescript-sdk-guidelines)** - Tinybird TypeScript SDK 使用指南
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">HashiCorp Terraform 团队的技能</h3></summary>
|
||
|
||
- **[hashicorp/azure-verified-modules](https://officialskills.sh/hashicorp/skills/azure-verified-modules)** - Terraform 模块的 Azure Verified Modules (AVM) 认证标准
|
||
- **[hashicorp/new-terraform-provider](https://officialskills.sh/hashicorp/skills/new-terraform-provider)** - 使用 Plugin Framework 搭建新的 Terraform provider 项目
|
||
- **[hashicorp/provider-resources](https://officialskills.sh/hashicorp/skills/provider-resources)** - 使用 Plugin Framework 实现 Terraform Provider 的 resources 与 data sources
|
||
- **[hashicorp/provider-test-patterns](https://officialskills.sh/hashicorp/skills/provider-test-patterns)** - 使用 terraform-plugin-testing 的 Terraform provider 验收测试模式
|
||
- **[hashicorp/provider-actions](https://officialskills.sh/hashicorp/skills/provider-actions)** - 使用 Plugin Framework 实现 Terraform Provider Actions
|
||
- **[hashicorp/run-acceptance-tests](https://officialskills.sh/hashicorp/skills/run-acceptance-tests)** - 使用 Go 测试运行器为 Terraform provider 运行验收测试
|
||
- **[hashicorp/refactor-module](https://officialskills.sh/hashicorp/skills/refactor-module)** - 将单体 Terraform 配置转换为可复用模块
|
||
- **[hashicorp/terraform-search-import](https://officialskills.sh/hashicorp/skills/terraform-search-import)** - 发现现有云资源并批量导入 Terraform state
|
||
- **[hashicorp/terraform-style-guide](https://officialskills.sh/hashicorp/skills/terraform-style-guide)** - 遵循 HashiCorp 官方风格约定生成 Terraform HCL 代码
|
||
- **[hashicorp/terraform-stacks](https://officialskills.sh/hashicorp/skills/terraform-stacks)** - 跨多个环境、区域与云账户管理基础设施
|
||
- **[hashicorp/terraform-test](https://officialskills.sh/hashicorp/skills/terraform-test)** - 使用 .tftest.hcl 文件的 Terraform 配置内置测试框架
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Sanity 团队的技能</h3></summary>
|
||
|
||
- **[sanity-io/sanity-best-practices](https://officialskills.sh/sanity-io/skills/sanity-best-practices)** - Sanity Studio、GROQ 查询与内容工作流的最佳实践
|
||
- **[sanity-io/content-modeling-best-practices](https://officialskills.sh/sanity-io/skills/content-modeling-best-practices)** - 在 Sanity 中设计可扩展内容模型的指南
|
||
- **[sanity-io/seo-aeo-best-practices](https://officialskills.sh/sanity-io/skills/seo-aeo-best-practices)** - 内容站点的 SEO 与答案引擎优化(AEO)模式
|
||
- **[sanity-io/content-experimentation-best-practices](https://officialskills.sh/sanity-io/skills/content-experimentation-best-practices)** - 内容 A/B 测试与实验工作流
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Firecrawl 团队的技能</h3></summary>
|
||
|
||
- **[firecrawl/firecrawl-build](https://officialskills.sh/firecrawl/skills/firecrawl-build)** - 将 Firecrawl 集成到应用代码中,用于网页搜索、抓取、提取和浏览器交互
|
||
- **[firecrawl/firecrawl-build-interact](https://officialskills.sh/firecrawl/skills/firecrawl-build-interact)** - 多步骤 Firecrawl 浏览器流程:点击、表单填写、分页和感知认证的导航
|
||
- **[firecrawl/firecrawl-build-onboarding](https://officialskills.sh/firecrawl/skills/firecrawl-build-onboarding)** - 在项目中首次集成时配置 Firecrawl 凭据和 SDK
|
||
- **[firecrawl/firecrawl-build-scrape](https://officialskills.sh/firecrawl/skills/firecrawl-build-scrape)** - 从产品代码中集成 Firecrawl `/scrape`,实现单页提取
|
||
- **[firecrawl/firecrawl-build-search](https://officialskills.sh/firecrawl/skills/firecrawl-build-search)** - 集成 Firecrawl `/search`,实现以查询为先的发现,并可选择性地填充内容
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Neon 的技能</h3></summary>
|
||
|
||
- **[neondatabase/neon-postgres](https://officialskills.sh/neondatabase/skills/neon-postgres)** - Neon Serverless Postgres 最佳实践
|
||
- **[neondatabase/claimable-postgres](https://officialskills.sh/neondatabase/skills/claimable-postgres)** - 使用 Neon 进行可申领(Claimable)Postgres 数据库配置
|
||
- **[neondatabase/neon-postgres-egress-optimizer](https://officialskills.sh/neondatabase/skills/neon-postgres-egress-optimizer)** - 优化 Neon Postgres 出站流量和数据传输
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">ClickHouse 的技能</h3></summary>
|
||
|
||
- **[clickhouse/clickhouse-best-practices](https://officialskills.sh/clickhouse/skills/clickhouse-best-practices)** - 使用 ClickHouse 的最佳实践
|
||
- **[clickhouse/chdb-datastore](https://officialskills.sh/clickhouse/skills/chdb-datastore)** - 可即插即用替代 pandas,在 16+ 数据源上获得 ClickHouse 性能
|
||
- **[clickhouse/chdb-sql](https://officialskills.sh/clickhouse/skills/chdb-sql)** - 面向 Python 的进程内 ClickHouse SQL 引擎 — 无需服务器即可查询文件、数据库和云存储
|
||
- **[clickhouse/clickhouse-architecture-advisor](https://officialskills.sh/clickhouse/skills/clickhouse-architecture-advisor)** - 设计 ClickHouse 架构,并将最佳实践转化为针对具体工作负载的决策
|
||
- **[clickhouse/clickhousectl-cloud-deploy](https://officialskills.sh/clickhouse/skills/clickhousectl-cloud-deploy)** - 使用 clickhousectl 部署到 ClickHouse Cloud,并从本地环境迁移
|
||
- **[clickhouse/clickhousectl-local-dev](https://officialskills.sh/clickhouse/skills/clickhousectl-local-dev)** - 使用 clickhousectl 从零搭建本地 ClickHouse 开发环境
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Remotion 的技能</h3></summary>
|
||
|
||
- **[remotion-dev/remotion](https://officialskills.sh/remotion-dev/skills/remotion)** - 使用 React 以编程方式创建视频
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Replicate 的技能</h3></summary>
|
||
|
||
- **[replicate/replicate](https://officialskills.sh/replicate/skills/replicate)** - 使用 Replicate 的 API 发现、比较并运行 AI 模型
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Typefully 的技能</h3></summary>
|
||
|
||
- **[typefully/typefully](https://officialskills.sh/typefully/skills/typefully)** - 在 X、LinkedIn、Threads、Bluesky 和 Mastodon 上创建、排期和发布社交媒体内容
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Venice.ai 的技能</h3></summary>
|
||
|
||
Venice.ai 为 Venice API 提供的官方技能。
|
||
|
||
- **[veniceai/venice-api-overview](https://github.com/veniceai/skills/tree/main/skills/venice-api-overview)** - API 基础、认证模式、定价和版本管理
|
||
- **[veniceai/venice-auth](https://github.com/veniceai/skills/tree/main/skills/venice-auth)** - API 密钥和基于钱包的 Venice 认证
|
||
- **[veniceai/venice-chat](https://github.com/veniceai/skills/tree/main/skills/venice-chat)** - 聊天补全、多模态输入、工具和流式传输
|
||
- **[veniceai/venice-responses](https://github.com/veniceai/skills/tree/main/skills/venice-responses)** - 适用于 Venice 的 OpenAI 兼容 Responses API
|
||
- **[veniceai/venice-embeddings](https://github.com/veniceai/skills/tree/main/skills/venice-embeddings)** - Embeddings 模型、维度和编码格式
|
||
- **[veniceai/venice-image-generate](https://github.com/veniceai/skills/tree/main/skills/venice-image-generate)** - 图像生成端点和可用风格
|
||
- **[veniceai/venice-image-edit](https://github.com/veniceai/skills/tree/main/skills/venice-image-edit)** - 图像编辑、放大和背景移除
|
||
- **[veniceai/venice-audio-speech](https://github.com/veniceai/skills/tree/main/skills/venice-audio-speech)** - 文本转语音模型、音色、格式和流式传输
|
||
- **[veniceai/venice-audio-music](https://github.com/veniceai/skills/tree/main/skills/venice-audio-music)** - 音乐生成排队、检索和完成端点
|
||
- **[veniceai/venice-audio-transcription](https://github.com/veniceai/skills/tree/main/skills/venice-audio-transcription)** - 音频转录模型和语音转文本选项
|
||
- **[veniceai/venice-video](https://github.com/veniceai/skills/tree/main/skills/venice-video)** - 视频生成和转录工作流
|
||
- **[veniceai/venice-models](https://github.com/veniceai/skills/tree/main/skills/venice-models)** - 模型目录、特性与兼容性映射
|
||
- **[veniceai/venice-characters](https://github.com/veniceai/skills/tree/main/skills/venice-characters)** - 角色端点和 `character_slug` 用法
|
||
- **[veniceai/venice-api-keys](https://github.com/veniceai/skills/tree/main/skills/venice-api-keys)** - API 密钥 CRUD、速率限制和 Web3 密钥
|
||
- **[veniceai/venice-billing](https://github.com/veniceai/skills/tree/main/skills/venice-billing)** - 余额、用量和计费分析端点
|
||
- **[veniceai/venice-x402](https://github.com/veniceai/skills/tree/main/skills/venice-x402)** - Base 上的钱包额度与 x402 支付
|
||
- **[veniceai/venice-crypto-rpc](https://github.com/veniceai/skills/tree/main/skills/venice-crypto-rpc)** - 为支持的加密网络提供 JSON-RPC 代理
|
||
- **[veniceai/venice-augment](https://github.com/veniceai/skills/tree/main/skills/venice-augment)** - 搜索、抓取和文本解析端点
|
||
- **[veniceai/venice-errors](https://github.com/veniceai/skills/tree/main/skills/venice-errors)** - 错误处理、重试和 API 状态码
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Vercel 工程团队的技能</h3></summary>
|
||
|
||
- **[vercel-labs/next-best-practices](https://officialskills.sh/vercel-labs/skills/next-best-practices)** - Next.js 最佳实践和推荐模式
|
||
- **[vercel-labs/next-cache-components](https://officialskills.sh/vercel-labs/skills/next-cache-components)** - Next.js 中的缓存策略和感知缓存的组件
|
||
- **[vercel-labs/next-upgrade](https://officialskills.sh/vercel-labs/skills/next-upgrade)** - 将 Next.js 项目升级到新版本
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Cloudflare 团队的技能</h3></summary>
|
||
|
||
- **[cloudflare/agents-sdk](https://officialskills.sh/cloudflare/skills/agents-sdk)** - 构建具有调度、RPC 和 MCP 服务器能力的有状态 AI Agent
|
||
- **[cloudflare/cloudflare](https://officialskills.sh/cloudflare/skills/cloudflare)** - 全面的 Cloudflare 平台技能,涵盖 Workers、Pages、存储、AI、网络、安全和 IaC
|
||
- **[cloudflare/cloudflare-email-service](https://officialskills.sh/cloudflare/skills/cloudflare-email-service)** - 使用 Cloudflare Email Sending 和 Email Routing 发送事务邮件并路由入站邮件
|
||
- **[cloudflare/durable-objects](https://officialskills.sh/cloudflare/skills/durable-objects)** - 通过 RPC、SQLite 和 WebSockets 实现有状态协调
|
||
- **[cloudflare/sandbox-sdk](https://officialskills.sh/cloudflare/skills/sandbox-sdk)** - 构建沙箱应用,在 Workers 上安全、隔离地执行代码
|
||
- **[cloudflare/web-perf](https://officialskills.sh/cloudflare/skills/web-perf)** - 审计 Core Web Vitals 和阻塞渲染的资源
|
||
- **[cloudflare/workers-best-practices](https://officialskills.sh/cloudflare/skills/workers-best-practices)** - 依据生产最佳实践和 wrangler.jsonc 约定审查和编写 Workers 代码
|
||
- **[cloudflare/wrangler](https://officialskills.sh/cloudflare/skills/wrangler)** - 部署和管理 Workers、KV、R2、D1、Vectorize、Queues、Workflows
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Netlify 团队的 Skills</h3></summary>
|
||
|
||
- **[netlify/netlify-functions](https://officialskills.sh/netlify/skills/netlify-functions)** - 构建无服务器 API 端点和后台任务
|
||
- **[netlify/netlify-edge-functions](https://officialskills.sh/netlify/skills/netlify-edge-functions)** - 低延迟边缘中间件和地理位置逻辑
|
||
- **[netlify/netlify-blobs](https://officialskills.sh/netlify/skills/netlify-blobs)** - 用于文件和数据的键值对象存储
|
||
- **[netlify/netlify-db](https://officialskills.sh/netlify/skills/netlify-db)** - 托管 Postgres,支持部署预览分支
|
||
- **[netlify/netlify-image-cdn](https://officialskills.sh/netlify/skills/netlify-image-cdn)** - 通过 CDN 优化和转换图像
|
||
- **[netlify/netlify-forms](https://officialskills.sh/netlify/skills/netlify-forms)** - HTML 表单处理,含垃圾邮件过滤
|
||
- **[netlify/netlify-frameworks](https://officialskills.sh/netlify/skills/netlify-frameworks)** - 部署支持 SSR 的 Web 框架
|
||
- **[netlify/netlify-caching](https://officialskills.sh/netlify/skills/netlify-caching)** - 配置 CDN 缓存和缓存清除
|
||
- **[netlify/netlify-config](https://officialskills.sh/netlify/skills/netlify-config)** - netlify.toml 站点配置参考
|
||
- **[netlify/netlify-cli-and-deploy](https://officialskills.sh/netlify/skills/netlify-cli-and-deploy)** - CLI 设置、本地开发与部署工作流
|
||
- **[netlify/netlify-deploy](https://officialskills.sh/netlify/skills/netlify-deploy)** - Netlify 站点的自动化部署工作流
|
||
- **[netlify/netlify-ai-gateway](https://officialskills.sh/netlify/skills/netlify-ai-gateway)** - 通过统一网关端点访问 AI 模型
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Google Labs(Stitch)的 Skills</h3></summary>
|
||
|
||
适用于 Stitch MCP 服务器的 Agent Skills,兼容 Claude Code、Gemini CLI、Cursor 等。
|
||
|
||
- **[google-labs-code/design-md](https://officialskills.sh/google-labs-code/skills/design-md)** - 创建和管理 DESIGN.md 文件
|
||
- **[google-labs-code/enhance-prompt](https://officialskills.sh/google-labs-code/skills/enhance-prompt)** - 利用设计规范和 UI/UX 词汇改进提示词
|
||
- **[google-labs-code/react-components](https://officialskills.sh/google-labs-code/skills/react-components)** - 将 Stitch 转换为 React 组件
|
||
- **[google-labs-code/remotion](https://officialskills.sh/google-labs-code/skills/remotion)** - 根据 Stitch 应用设计生成演示视频
|
||
- **[google-labs-code/shadcn-ui](https://officialskills.sh/google-labs-code/skills/shadcn-ui)** - 使用 shadcn/ui 构建 UI 组件
|
||
- **[google-labs-code/stitch-loop](https://officialskills.sh/google-labs-code/skills/stitch-loop)** - 迭代式设计到代码反馈循环
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Google Workspace CLI 的 Skills</h3></summary>
|
||
|
||
官方 Google Workspace CLI 技能,用于通过 `gws` CLI 工具管理 Google Workspace 服务。
|
||
|
||
- **[googleworkspace/gws-shared](https://officialskills.sh/googleworkspace/skills/gws-shared)** - 共享认证、全局标志和输出格式化
|
||
- **[googleworkspace/gws-drive](https://officialskills.sh/googleworkspace/skills/gws-drive)** - 管理 Google Drive 文件、文件夹和共享云端硬盘
|
||
- **[googleworkspace/gws-sheets](https://officialskills.sh/googleworkspace/skills/gws-sheets)** - 读取和写入 Google Sheets 电子表格
|
||
- **[googleworkspace/gws-gmail](https://officialskills.sh/googleworkspace/skills/gws-gmail)** - 发送、读取和管理 Gmail 邮件
|
||
- **[googleworkspace/gws-calendar](https://officialskills.sh/googleworkspace/skills/gws-calendar)** - 管理 Google Calendar 日历和事件
|
||
- **[googleworkspace/gws-admin-reports](https://officialskills.sh/googleworkspace/skills/gws-admin-reports)** - Workspace 审计日志和使用报告
|
||
- **[googleworkspace/gws-docs](https://officialskills.sh/googleworkspace/skills/gws-docs)** - 读取和写入 Google Docs 文档
|
||
- **[googleworkspace/gws-slides](https://officialskills.sh/googleworkspace/skills/gws-slides)** - 读取和写入 Google Slides 演示文稿
|
||
- **[googleworkspace/gws-tasks](https://officialskills.sh/googleworkspace/skills/gws-tasks)** - 管理 Google Tasks 任务列表和任务
|
||
- **[googleworkspace/gws-people](https://officialskills.sh/googleworkspace/skills/gws-people)** - 管理 Google People 联系人和个人资料
|
||
- **[googleworkspace/gws-chat](https://officialskills.sh/googleworkspace/skills/gws-chat)** - 管理 Google Chat 空间和消息
|
||
- **[googleworkspace/gws-classroom](https://officialskills.sh/googleworkspace/skills/gws-classroom)** - 管理 Google Classroom 班级、花名册和课程作业
|
||
- **[googleworkspace/gws-forms](https://officialskills.sh/googleworkspace/skills/gws-forms)** - 读取和写入 Google Forms
|
||
- **[googleworkspace/gws-keep](https://officialskills.sh/googleworkspace/skills/gws-keep)** - 管理 Google Keep 笔记
|
||
- **[googleworkspace/gws-events](https://officialskills.sh/googleworkspace/skills/gws-events)** - 订阅 Google Workspace 事件
|
||
- **[googleworkspace/gws-modelarmor](https://officialskills.sh/googleworkspace/skills/gws-modelarmor)** - 过滤用户生成内容以确保安全
|
||
- **[googleworkspace/gws-workflow](https://officialskills.sh/googleworkspace/skills/gws-workflow)** - 跨服务的 Google Workspace 生产力工作流
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Expo 团队的 Skills</h3></summary>
|
||
|
||
Expo 团队提供的官方 AI 代理技能,用于构建、部署和调试 Expo 应用。
|
||
|
||
- **[expo/building-native-ui](https://officialskills.sh/expo/skills/building-native-ui)** - 使用 Expo Router、样式、组件、导航和动画构建应用
|
||
- **[expo/expo-api-routes](https://officialskills.sh/expo/skills/expo-api-routes)** - 在 Expo Router 中使用 EAS Hosting 创建 API 路由
|
||
- **[expo/expo-cicd-workflows](https://officialskills.sh/expo/skills/expo-cicd-workflows)** - Expo 项目的 CI/CD 工作流
|
||
- **[expo/expo-deployment](https://officialskills.sh/expo/skills/expo-deployment)** - 将 Expo 应用部署到生产环境
|
||
- **[expo/expo-dev-client](https://officialskills.sh/expo/skills/expo-dev-client)** - 在本地或通过 TestFlight 构建和分发 Expo 开发客户端
|
||
- **[expo/expo-tailwind-setup](https://officialskills.sh/expo/skills/expo-tailwind-setup)** - 在 Expo 中使用 NativeWind v5 设置 Tailwind CSS v4
|
||
- **[expo/expo-ui-jetpack-compose](https://officialskills.sh/expo/skills/expo-ui-jetpack-compose)** - 适用于 Expo 的 Jetpack Compose UI 组件
|
||
- **[expo/expo-ui-swift-ui](https://officialskills.sh/expo/skills/expo-ui-swift-ui)** - 适用于 Expo 的 SwiftUI 组件
|
||
- **[expo/native-data-fetching](https://officialskills.sh/expo/skills/native-data-fetching)** - 网络请求、API 调用、缓存和离线支持
|
||
- **[expo/upgrading-expo](https://officialskills.sh/expo/skills/upgrading-expo)** - 升级 Expo SDK 版本
|
||
- **[expo/use-dom](https://officialskills.sh/expo/skills/use-dom)** - 使用 DOM 组件在原生 WebView 中运行 Web 代码
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Hugging Face 团队的 Skills</h3></summary>
|
||
|
||
Hugging Face 团队提供的官方 AI 代理技能,用于机器学习(ML)工作流。
|
||
|
||
- **[huggingface/hf-cli](https://officialskills.sh/huggingface/skills/hf-cli)** - 用于 Hub 操作的 HF CLI 工具
|
||
- **[huggingface/hugging-face-dataset-viewer](https://officialskills.sh/huggingface/skills/hugging-face-dataset-viewer)** - 使用 Dataset Viewer API 浏览和查询 HF 数据集
|
||
- **[huggingface/hugging-face-datasets](https://officialskills.sh/huggingface/skills/hugging-face-datasets)** - 创建和管理数据集,支持配置和 SQL 查询
|
||
- **[huggingface/hugging-face-evaluation](https://officialskills.sh/huggingface/skills/hugging-face-evaluation)** - 使用 vLLM/lighteval 进行模型评估和评估表
|
||
- **[huggingface/hugging-face-jobs](https://officialskills.sh/huggingface/skills/hugging-face-jobs)** - 在 HF 基础设施上运行计算任务和 Python 脚本
|
||
- **[huggingface/hugging-face-model-trainer](https://officialskills.sh/huggingface/skills/hugging-face-model-trainer)** - 使用 TRL 训练模型:SFT、DPO、GRPO、GGUF 转换
|
||
- **[huggingface/hugging-face-paper-pages](https://officialskills.sh/huggingface/skills/hugging-face-paper-pages)** - 在 HF Hub 上创建和管理论文页面
|
||
- **[huggingface/hugging-face-paper-publisher](https://officialskills.sh/huggingface/skills/hugging-face-paper-publisher)** - 在 HF Hub 上发布论文并关联模型/数据集链接
|
||
- **[huggingface/hugging-face-tool-builder](https://officialskills.sh/huggingface/skills/hugging-face-tool-builder)** - 构建可复用的 HF API 操作脚本
|
||
- **[huggingface/hugging-face-trackio](https://officialskills.sh/huggingface/skills/hugging-face-trackio)** - 使用实时仪表板跟踪 ML 实验
|
||
- **[huggingface/hugging-face-vision-trainer](https://officialskills.sh/huggingface/skills/hugging-face-vision-trainer)** - 在 HF 基础设施上训练视觉模型
|
||
- **[huggingface/huggingface-gradio](https://officialskills.sh/huggingface/skills/huggingface-gradio)** - 构建 Gradio 应用并部署到 HF Spaces
|
||
- **[huggingface/transformers.js](https://officialskills.sh/huggingface/skills/transformers.js)** - 使用 Transformers.js 在浏览器中运行 ML 模型
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Trail of Bits 团队的安全技能</h3></summary>
|
||
|
||
- **[trailofbits/ask-questions-if-underspecified](https://officialskills.sh/trailofbits/skills/ask-questions-if-underspecified)** - 针对模糊需求提示进行澄清
|
||
- **[trailofbits/audit-context-building](https://officialskills.sh/trailofbits/skills/audit-context-building)** - 通过超细粒度代码分析构建深度架构上下文
|
||
- **[trailofbits/building-secure-contracts](https://officialskills.sh/trailofbits/skills/building-secure-contracts)** - 智能合约安全工具包,为 6 条区块链提供漏洞扫描器
|
||
- **[trailofbits/burpsuite-project-parser](https://officialskills.sh/trailofbits/skills/burpsuite-project-parser)** - 从 Burp Suite 项目文件中搜索并提取数据
|
||
- **[trailofbits/claude-in-chrome-troubleshooting](https://officialskills.sh/trailofbits/skills/claude-in-chrome-troubleshooting)** - 诊断并修复 Claude in Chrome MCP 扩展的连接问题
|
||
- **[trailofbits/constant-time-analysis](https://officialskills.sh/trailofbits/skills/constant-time-analysis)** - 检测加密代码中由编译器引发的定时侧信道
|
||
- **[trailofbits/culture-index](https://officialskills.sh/trailofbits/skills/culture-index)** - 索引并搜索文化文档
|
||
- **[trailofbits/differential-review](https://officialskills.sh/trailofbits/skills/differential-review)** - 面向安全的 diff 审查,结合 git 历史分析
|
||
- **[trailofbits/dwarf-expert](https://officialskills.sh/trailofbits/skills/dwarf-expert)** - DWARF 调试格式专业知识
|
||
- **[trailofbits/entry-point-analyzer](https://officialskills.sh/trailofbits/skills/entry-point-analyzer)** - 识别智能合约中会改变状态的入口点
|
||
- **[trailofbits/firebase-apk-scanner](https://officialskills.sh/trailofbits/skills/firebase-apk-scanner)** - 扫描 Android APK,检测 Firebase 误配置和安全漏洞
|
||
- **[trailofbits/insecure-defaults](https://officialskills.sh/trailofbits/skills/insecure-defaults)** - 检测不安全的默认配置,如硬编码密钥、默认凭据和弱加密
|
||
- **[trailofbits/modern-python](https://officialskills.sh/trailofbits/skills/modern-python)** - 现代 Python 工具链,涵盖 uv、ruff、ty 和 pytest 最佳实践
|
||
- **[trailofbits/property-based-testing](https://officialskills.sh/trailofbits/skills/property-based-testing)** - 面向多种语言和智能合约的属性测试(property-based testing)
|
||
- **[trailofbits/semgrep-rule-creator](https://officialskills.sh/trailofbits/skills/semgrep-rule-creator)** - 创建并优化用于漏洞检测的 Semgrep 规则
|
||
- **[trailofbits/semgrep-rule-variant-creator](https://officialskills.sh/trailofbits/skills/semgrep-rule-variant-creator)** - 将现有 Semgrep 规则移植到新目标语言,并进行测试驱动验证
|
||
- **[trailofbits/sharp-edges](https://officialskills.sh/trailofbits/skills/sharp-edges)** - 识别易出错的 API 和危险配置
|
||
- **[trailofbits/spec-to-code-compliance](https://officialskills.sh/trailofbits/skills/spec-to-code-compliance)** - 面向区块链审计的规格到代码合规检查器
|
||
- **[trailofbits/static-analysis](https://officialskills.sh/trailofbits/skills/static-analysis)** - 静态分析工具包,集成 CodeQL、Semgrep 和 SARIF
|
||
- **[trailofbits/testing-handbook-skills](https://officialskills.sh/trailofbits/skills/testing-handbook-skills)** - Testing Handbook 技能:模糊测试器、静态分析、消毒器(sanitizers)
|
||
- **[trailofbits/variant-analysis](https://officialskills.sh/trailofbits/skills/variant-analysis)** - 通过基于模式的分析发现类似漏洞
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Sentry 团队为其开发团队提供的技能</h3></summary>
|
||
|
||
- **[getsentry/sentry-sdk-setup](https://officialskills.sh/getsentry/skills/sentry-sdk-setup)** - 在任何语言或框架中设置 Sentry——自动检测平台并路由到正确的 SDK
|
||
- **[getsentry/sentry-workflow](https://officialskills.sh/getsentry/skills/sentry-workflow)** - 端到端 Sentry 工作流:借助 Sentry 上下文修复生产问题并审查代码
|
||
- **[getsentry/sentry-fix-issues](https://officialskills.sh/getsentry/skills/sentry-fix-issues)** - 通过 MCP 借助堆栈跟踪、breadcrumb 和 trace 上下文查找并修复 Sentry 问题
|
||
- **[getsentry/sentry-code-review](https://officialskills.sh/getsentry/skills/sentry-code-review)** - 使用 Sentry 问题和 trace 上下文审查代码变更
|
||
- **[getsentry/sentry-pr-code-review](https://officialskills.sh/getsentry/skills/sentry-pr-code-review)** - 审查来自 Seer Bug Prediction 和 Sentry 反馈的 PR 评论
|
||
- **[getsentry/sentry-create-alert](https://officialskills.sh/getsentry/skills/sentry-create-alert)** - 创建 Sentry 告警,支持 email、Slack、PagerDuty、Discord 等
|
||
- **[getsentry/sentry-feature-setup](https://officialskills.sh/getsentry/skills/sentry-feature-setup)** - 配置高级 Sentry 功能:AI 监控、OTel 管道和告警
|
||
- **[getsentry/sentry-otel-exporter-setup](https://officialskills.sh/getsentry/skills/sentry-otel-exporter-setup)** - 配置带有 Sentry Exporter 的 OpenTelemetry Collector
|
||
- **[getsentry/sentry-setup-ai-monitoring](https://officialskills.sh/getsentry/skills/sentry-setup-ai-monitoring)** - 为 OpenAI、Anthropic、Vercel AI、LangChain、Google GenAI 和 Pydantic AI 接入监控
|
||
- **[getsentry/sentry-sdk-upgrade](https://officialskills.sh/getsentry/skills/sentry-sdk-upgrade)** - 跨主要版本升级 Sentry JavaScript SDK
|
||
- **[getsentry/sentry-sdk-skill-creator](https://officialskills.sh/getsentry/skills/sentry-sdk-skill-creator)** - 为某个平台创建新的 Sentry SDK 技能包
|
||
- **[getsentry/sentry-android-sdk](https://officialskills.sh/getsentry/skills/sentry-android-sdk)** - Android 完整 Sentry SDK 设置(Kotlin 和 Java)
|
||
- **[getsentry/sentry-browser-sdk](https://officialskills.sh/getsentry/skills/sentry-browser-sdk)** - 浏览器 JavaScript 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-cloudflare-sdk](https://officialskills.sh/getsentry/skills/sentry-cloudflare-sdk)** - Cloudflare Workers、Pages、Durable Objects、Queues 和 Workflows 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-cocoa-sdk](https://officialskills.sh/getsentry/skills/sentry-cocoa-sdk)** - Apple 平台完整 Sentry SDK 设置(iOS、macOS、tvOS、watchOS、visionOS)
|
||
- **[getsentry/sentry-dotnet-sdk](https://officialskills.sh/getsentry/skills/sentry-dotnet-sdk)** - .NET 完整 Sentry SDK 设置(ASP.NET Core、MAUI、WPF、WinForms、Blazor、Azure Functions)
|
||
- **[getsentry/sentry-elixir-sdk](https://officialskills.sh/getsentry/skills/sentry-elixir-sdk)** - Elixir 完整 Sentry SDK 设置,支持 Phoenix、Plug、LiveView、Oban 和 Quantum
|
||
- **[getsentry/sentry-flutter-sdk](https://officialskills.sh/getsentry/skills/sentry-flutter-sdk)** - 全平台 Flutter 和 Dart 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-go-sdk](https://officialskills.sh/getsentry/skills/sentry-go-sdk)** - Go 完整 Sentry SDK 设置(net/http、Gin、Echo、Fiber、FastHTTP、Iris、Negroni)
|
||
- **[getsentry/sentry-nestjs-sdk](https://officialskills.sh/getsentry/skills/sentry-nestjs-sdk)** - NestJS 完整 Sentry SDK 设置,支持 Express 或 Fastify、GraphQL、微服务
|
||
- **[getsentry/sentry-nextjs-sdk](https://officialskills.sh/getsentry/skills/sentry-nextjs-sdk)** - Next.js 13+ 完整 Sentry SDK 设置(App Router 和 Pages Router)
|
||
- **[getsentry/sentry-node-sdk](https://officialskills.sh/getsentry/skills/sentry-node-sdk)** - Node.js、Bun 和 Deno 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-php-sdk](https://officialskills.sh/getsentry/skills/sentry-php-sdk)** - PHP、Laravel 和 Symfony 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-python-sdk](https://officialskills.sh/getsentry/skills/sentry-python-sdk)** - Python 完整 Sentry SDK 设置(Django、Flask、FastAPI、Celery、Starlette、AIOHTTP、Tornado)
|
||
- **[getsentry/sentry-react-native-sdk](https://officialskills.sh/getsentry/skills/sentry-react-native-sdk)** - React Native 和 Expo 完整 Sentry SDK 设置
|
||
- **[getsentry/sentry-react-sdk](https://officialskills.sh/getsentry/skills/sentry-react-sdk)** - React 完整 Sentry SDK 设置(React Router v5-v7、TanStack Router、Redux、Vite、webpack)
|
||
- **[getsentry/sentry-ruby-sdk](https://officialskills.sh/getsentry/skills/sentry-ruby-sdk)** - Ruby 完整 Sentry SDK 设置(Rails、Sinatra、Rack、Sidekiq、Resque)
|
||
- **[getsentry/sentry-svelte-sdk](https://officialskills.sh/getsentry/skills/sentry-svelte-sdk)** - Svelte 和 SvelteKit 完整 Sentry SDK 设置
|
||
|
||
</details>
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Microsoft 提供的 Skills</h3></summary>
|
||
|
||
面向 Azure SDK 与 Microsoft AI Foundry 开发的领域专属知识。6 种语言共 133 个 skills。
|
||
|
||
### 核心 Skills
|
||
|
||
- **[microsoft/cloud-solution-architect](https://officialskills.sh/microsoft/skills/cloud-solution-architect)** - 设计良好架构的 Azure 云系统
|
||
- **[microsoft/continual-learning](https://officialskills.sh/microsoft/skills/continual-learning)** - Azure AI 的持续学习(Continual Learning)模式
|
||
- **[microsoft/copilot-sdk](https://officialskills.sh/microsoft/skills/copilot-sdk)** - 基于 GitHub Copilot SDK 构建应用
|
||
- **[microsoft/entra-agent-id](https://officialskills.sh/microsoft/skills/entra-agent-id)** - 通过 Graph API 使用 Microsoft Entra Agent ID OAuth2 身份
|
||
- **[microsoft/frontend-design-review](https://officialskills.sh/microsoft/skills/frontend-design-review)** - 评审并打造独具特色的前端界面
|
||
- **[microsoft/github-issue-creator](https://officialskills.sh/microsoft/skills/github-issue-creator)** - 将笔记整理为结构化的 GitHub issue 报告
|
||
- **[microsoft/mcp-builder](https://officialskills.sh/microsoft/skills/mcp-builder)** - 面向 LLM 工具集成的 MCP 服务器创建指南
|
||
- **[microsoft/podcast-generation](https://officialskills.sh/microsoft/skills/podcast-generation)** - 使用 Azure OpenAI Realtime API 生成 AI 播客音频
|
||
- **[microsoft/skill-creator](https://officialskills.sh/microsoft/skills/skill-creator)** - 为 AI 编程代理创建高效 skills 的指南
|
||
|
||
### .NET Skills
|
||
|
||
- **[microsoft/azure-ai-document-intelligence-dotnet](https://officialskills.sh/microsoft/skills/azure-ai-document-intelligence-dotnet)** - 文档文本、表格与数据提取
|
||
- **[microsoft/azure-ai-openai-dotnet](https://officialskills.sh/microsoft/skills/azure-ai-openai-dotnet)** - GPT-4、embeddings、DALL-E 与 Whisper 客户端
|
||
- **[microsoft/azure-ai-projects-dotnet](https://officialskills.sh/microsoft/skills/azure-ai-projects-dotnet)** - AI Foundry 项目管理 SDK
|
||
- **[microsoft/azure-ai-voicelive-dotnet](https://officialskills.sh/microsoft/skills/azure-ai-voicelive-dotnet)** - 实时双向语音 AI
|
||
- **[microsoft/azure-eventgrid-dotnet](https://officialskills.sh/microsoft/skills/azure-eventgrid-dotnet)** - Event Grid 主题与域发布
|
||
- **[microsoft/azure-eventhub-dotnet](https://officialskills.sh/microsoft/skills/azure-eventhub-dotnet)** - 高吞吐量事件流
|
||
- **[microsoft/azure-identity-dotnet](https://officialskills.sh/microsoft/skills/azure-identity-dotnet)** - Microsoft Entra ID 身份验证
|
||
- **[microsoft/azure-maps-search-dotnet](https://officialskills.sh/microsoft/skills/azure-maps-search-dotnet)** - 地理编码、路线规划与天气服务
|
||
- **[microsoft/azure-mgmt-apicenter-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-apicenter-dotnet)** - API 清单与治理
|
||
- **[microsoft/azure-mgmt-apimanagement-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-apimanagement-dotnet)** - 通过 ARM 配置 API Management
|
||
- **[microsoft/azure-mgmt-applicationinsights-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-applicationinsights-dotnet)** - Application Insights 资源管理
|
||
- **[microsoft/azure-mgmt-arizeaiobservabilityeval-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-arizeaiobservabilityeval-dotnet)** - Arize AI 可观测性管理
|
||
- **[microsoft/azure-mgmt-botservice-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-botservice-dotnet)** - 通过 ARM 配置 Bot Service
|
||
- **[microsoft/azure-mgmt-fabric-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-fabric-dotnet)** - Microsoft Fabric 容量管理
|
||
- **[microsoft/azure-mgmt-mongodbatlas-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-mongodbatlas-dotnet)** - 将 MongoDB Atlas 作为 ARM 资源管理
|
||
- **[microsoft/azure-mgmt-weightsandbiases-dotnet](https://officialskills.sh/microsoft/skills/azure-mgmt-weightsandbiases-dotnet)** - Weights & Biases 部署管理
|
||
- **[microsoft/azure-resource-manager-cosmosdb-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-cosmosdb-dotnet)** - Cosmos DB 资源配置
|
||
- **[microsoft/azure-resource-manager-durabletask-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-durabletask-dotnet)** - Durable Task Scheduler 管理
|
||
- **[microsoft/azure-resource-manager-mysql-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-mysql-dotnet)** - MySQL Flexible Server 管理
|
||
- **[microsoft/azure-resource-manager-playwright-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-playwright-dotnet)** - Playwright Testing 工作区管理
|
||
- **[microsoft/azure-resource-manager-postgresql-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-postgresql-dotnet)** - PostgreSQL Flexible Server 管理
|
||
- **[microsoft/azure-resource-manager-redis-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-redis-dotnet)** - Azure Cache for Redis 资源配置
|
||
- **[microsoft/azure-resource-manager-sql-dotnet](https://officialskills.sh/microsoft/skills/azure-resource-manager-sql-dotnet)** - Azure SQL 资源管理
|
||
- **[microsoft/azure-search-documents-dotnet](https://officialskills.sh/microsoft/skills/azure-search-documents-dotnet)** - 全文、向量与混合搜索
|
||
- **[microsoft/azure-security-keyvault-keys-dotnet](https://officialskills.sh/microsoft/skills/azure-security-keyvault-keys-dotnet)** - 加密密钥管理
|
||
- **[microsoft/azure-servicebus-dotnet](https://officialskills.sh/microsoft/skills/azure-servicebus-dotnet)** - 基于队列与主题的企业级消息传递
|
||
- **[microsoft/m365-agents-dotnet](https://officialskills.sh/microsoft/skills/m365-agents-dotnet)** - M365、Teams 与 Copilot Studio 代理
|
||
- **[microsoft/microsoft-azure-webjobs-extensions-authentication-events-dotnet](https://officialskills.sh/microsoft/skills/microsoft-azure-webjobs-extensions-authentication-events-dotnet)** - Entra ID 自定义身份验证事件处理程序
|
||
|
||
### Java Skills
|
||
|
||
- **[microsoft/azure-ai-anomalydetector-java](https://officialskills.sh/microsoft/skills/azure-ai-anomalydetector-java)** - 异常检测应用
|
||
- **[microsoft/azure-ai-contentsafety-java](https://officialskills.sh/microsoft/skills/azure-ai-contentsafety-java)** - 内容审核与安全
|
||
- **[microsoft/azure-ai-formrecognizer-java](https://officialskills.sh/microsoft/skills/azure-ai-formrecognizer-java)** - 文档分析与表单提取
|
||
- **[microsoft/azure-ai-projects-java](https://officialskills.sh/microsoft/skills/azure-ai-projects-java)** - AI Foundry 项目管理
|
||
- **[microsoft/azure-ai-vision-imageanalysis-java](https://officialskills.sh/microsoft/skills/azure-ai-vision-imageanalysis-java)** - 图像描述、OCR 与目标检测
|
||
- **[microsoft/azure-ai-voicelive-java](https://officialskills.sh/microsoft/skills/azure-ai-voicelive-java)** - 实时双向语音 AI
|
||
- **[microsoft/azure-appconfiguration-java](https://officialskills.sh/microsoft/skills/azure-appconfiguration-java)** - 集中式应用配置管理
|
||
- **[microsoft/azure-communication-callautomation-java](https://officialskills.sh/microsoft/skills/azure-communication-callautomation-java)** - 带 IVR 与 AI 的通话自动化
|
||
- **[microsoft/azure-communication-callingserver-java](https://officialskills.sh/microsoft/skills/azure-communication-callingserver-java)** - CallingServer 旧版 SDK
|
||
- **[microsoft/azure-communication-chat-java](https://officialskills.sh/microsoft/skills/azure-communication-chat-java)** - 支持线程与回执的实时聊天
|
||
- **[microsoft/azure-communication-common-java](https://officialskills.sh/microsoft/skills/azure-communication-common-java)** - Communication Services 通用工具
|
||
- **[microsoft/azure-communication-sms-java](https://officialskills.sh/microsoft/skills/azure-communication-sms-java)** - 短信发送与送达报告
|
||
- **[microsoft/azure-compute-batch-java](https://officialskills.sh/microsoft/skills/azure-compute-batch-java)** - 大规模并行与 HPC 批处理作业
|
||
- **[microsoft/azure-cosmos-java](https://officialskills.sh/microsoft/skills/azure-cosmos-java)** - 支持全球分布的 Cosmos DB NoSQL
|
||
- **[microsoft/azure-data-tables-java](https://officialskills.sh/microsoft/skills/azure-data-tables-java)** - NoSQL 键值表存储
|
||
- **[microsoft/azure-eventgrid-java](https://officialskills.sh/microsoft/skills/azure-eventgrid-java)** - 事件驱动的发布/订阅消息传递
|
||
- **[microsoft/azure-eventhub-java](https://officialskills.sh/microsoft/skills/azure-eventhub-java)** - 实时高吞吐量流式处理
|
||
- **[microsoft/azure-identity-java](https://officialskills.sh/microsoft/skills/azure-identity-java)** - Microsoft Entra ID 身份验证
|
||
- **[microsoft/azure-messaging-webpubsub-java](https://officialskills.sh/microsoft/skills/azure-messaging-webpubsub-java)** - 实时 WebSocket 消息传递
|
||
- **[microsoft/azure-monitor-ingestion-java](https://officialskills.sh/microsoft/skills/azure-monitor-ingestion-java)** - 将自定义日志摄入 Azure Monitor
|
||
- **[microsoft/azure-monitor-opentelemetry-exporter-java](https://officialskills.sh/microsoft/skills/azure-monitor-opentelemetry-exporter-java)** - 将 OpenTelemetry 导出至 Azure Monitor
|
||
- **[microsoft/azure-monitor-query-java](https://officialskills.sh/microsoft/skills/azure-monitor-query-java)** - 查询 Azure Monitor 日志与指标
|
||
- **[microsoft/azure-security-keyvault-keys-java](https://officialskills.sh/microsoft/skills/azure-security-keyvault-keys-java)** - 加密密钥管理
|
||
- **[microsoft/azure-security-keyvault-secrets-java](https://officialskills.sh/microsoft/skills/azure-security-keyvault-secrets-java)** - 密码与密钥的机密管理
|
||
- **[microsoft/azure-storage-blob-java](https://officialskills.sh/microsoft/skills/azure-storage-blob-java)** - 用于文件管理的 Blob 存储
|
||
|
||
### Python 技能
|
||
|
||
- **[microsoft/agent-framework-azure-ai-py](https://officialskills.sh/microsoft/skills/agent-framework-azure-ai-py)** - 面向 Azure AI Foundry 的 Agent Framework(智能体框架)
|
||
- **[microsoft/agents-v2-py](https://officialskills.sh/microsoft/skills/agents-v2-py)** - Foundry Agents SDK — 基于容器的自定义镜像智能体
|
||
- **[microsoft/azure-ai-contentsafety-py](https://officialskills.sh/microsoft/skills/azure-ai-contentsafety-py)** - 有害内容检测
|
||
- **[microsoft/azure-ai-contentunderstanding-py](https://officialskills.sh/microsoft/skills/azure-ai-contentunderstanding-py)** - 多模态内容提取
|
||
- **[microsoft/azure-ai-ml-py](https://officialskills.sh/microsoft/skills/azure-ai-ml-py)** - Azure ML 工作区与作业管理
|
||
- **[microsoft/azure-ai-projects-py](https://officialskills.sh/microsoft/skills/azure-ai-projects-py)** - AI Foundry 项目客户端与智能体
|
||
- **[microsoft/azure-ai-textanalytics-py](https://officialskills.sh/microsoft/skills/azure-ai-textanalytics-py)** - NLP:情感分析、实体识别、关键短语
|
||
- **[microsoft/azure-ai-transcription-py](https://officialskills.sh/microsoft/skills/azure-ai-transcription-py)** - 语音转文字转录
|
||
- **[microsoft/azure-ai-translation-document-py](https://officialskills.sh/microsoft/skills/azure-ai-translation-document-py)** - 批量文档翻译
|
||
- **[microsoft/azure-ai-translation-text-py](https://officialskills.sh/microsoft/skills/azure-ai-translation-text-py)** - 实时文本翻译
|
||
- **[microsoft/azure-ai-vision-imageanalysis-py](https://officialskills.sh/microsoft/skills/azure-ai-vision-imageanalysis-py)** - 图像字幕、标签、OCR、对象检测
|
||
- **[microsoft/azure-ai-voicelive-py](https://officialskills.sh/microsoft/skills/azure-ai-voicelive-py)** - 实时双向语音 AI
|
||
- **[microsoft/azure-appconfiguration-py](https://officialskills.sh/microsoft/skills/azure-appconfiguration-py)** - 功能开关与动态配置
|
||
- **[microsoft/azure-containerregistry-py](https://officialskills.sh/microsoft/skills/azure-containerregistry-py)** - 容器镜像与注册表管理
|
||
- **[microsoft/azure-cosmos-db-py](https://officialskills.sh/microsoft/skills/azure-cosmos-db-py)** - Cosmos DB 配合 Python/FastAPI 模式
|
||
- **[microsoft/azure-cosmos-py](https://officialskills.sh/microsoft/skills/azure-cosmos-py)** - Cosmos DB NoSQL 客户端库
|
||
- **[microsoft/azure-data-tables-py](https://officialskills.sh/microsoft/skills/azure-data-tables-py)** - NoSQL 键值表存储
|
||
- **[microsoft/azure-eventgrid-py](https://officialskills.sh/microsoft/skills/azure-eventgrid-py)** - 事件驱动的发布/订阅路由
|
||
- **[microsoft/azure-eventhub-py](https://officialskills.sh/microsoft/skills/azure-eventhub-py)** - 高吞吐量事件流
|
||
- **[microsoft/azure-identity-py](https://officialskills.sh/microsoft/skills/azure-identity-py)** - Microsoft Entra ID 身份验证
|
||
- **[microsoft/azure-keyvault-py](https://officialskills.sh/microsoft/skills/azure-keyvault-py)** - 机密、密钥与证书管理
|
||
- **[microsoft/azure-messaging-webpubsubservice-py](https://officialskills.sh/microsoft/skills/azure-messaging-webpubsubservice-py)** - 实时 WebSocket 消息传递
|
||
- **[microsoft/azure-mgmt-apicenter-py](https://officialskills.sh/microsoft/skills/azure-mgmt-apicenter-py)** - API 清单与治理
|
||
- **[microsoft/azure-mgmt-apimanagement-py](https://officialskills.sh/microsoft/skills/azure-mgmt-apimanagement-py)** - API Management 服务管理
|
||
- **[microsoft/azure-mgmt-botservice-py](https://officialskills.sh/microsoft/skills/azure-mgmt-botservice-py)** - Bot Service 资源管理
|
||
- **[microsoft/azure-mgmt-fabric-py](https://officialskills.sh/microsoft/skills/azure-mgmt-fabric-py)** - Microsoft Fabric 容量管理
|
||
- **[microsoft/azure-monitor-ingestion-py](https://officialskills.sh/microsoft/skills/azure-monitor-ingestion-py)** - 自定义日志摄入至 Azure Monitor
|
||
- **[microsoft/azure-monitor-opentelemetry-exporter-py](https://officialskills.sh/microsoft/skills/azure-monitor-opentelemetry-exporter-py)** - OpenTelemetry 导出至 Application Insights
|
||
- **[microsoft/azure-monitor-opentelemetry-py](https://officialskills.sh/microsoft/skills/azure-monitor-opentelemetry-py)** - 一行代码完成 Application Insights 配置
|
||
- **[microsoft/azure-monitor-query-py](https://officialskills.sh/microsoft/skills/azure-monitor-query-py)** - 查询 Azure Monitor 日志与指标
|
||
- **[microsoft/azure-search-documents-py](https://officialskills.sh/microsoft/skills/azure-search-documents-py)** - 全文、向量与混合搜索
|
||
- **[microsoft/azure-servicebus-py](https://officialskills.sh/microsoft/skills/azure-servicebus-py)** - 企业级消息传递(队列与主题)
|
||
- **[microsoft/azure-speech-to-text-rest-py](https://officialskills.sh/microsoft/skills/azure-speech-to-text-rest-py)** - 短音频 REST 语音转文字
|
||
- **[microsoft/azure-storage-blob-py](https://officialskills.sh/microsoft/skills/azure-storage-blob-py)** - Blob 对象存储客户端
|
||
- **[microsoft/azure-storage-file-datalake-py](https://officialskills.sh/microsoft/skills/azure-storage-file-datalake-py)** - 分层数据湖存储
|
||
- **[microsoft/azure-storage-file-share-py](https://officialskills.sh/microsoft/skills/azure-storage-file-share-py)** - SMB 文件共享管理
|
||
- **[microsoft/azure-storage-queue-py](https://officialskills.sh/microsoft/skills/azure-storage-queue-py)** - 简单消息队列
|
||
- **[microsoft/fastapi-router-py](https://officialskills.sh/microsoft/skills/fastapi-router-py)** - 带 CRUD 与身份验证的 FastAPI 路由
|
||
- **[microsoft/m365-agents-py](https://officialskills.sh/microsoft/skills/m365-agents-py)** - M365、Teams 与 Copilot Studio 智能体
|
||
- **[microsoft/pydantic-models-py](https://officialskills.sh/microsoft/skills/pydantic-models-py)** - 用于 API 模式的 Pydantic 模型
|
||
|
||
### Rust 技能
|
||
|
||
- **[microsoft/azure-cosmos-rust](https://officialskills.sh/microsoft/skills/azure-cosmos-rust)** - Cosmos DB NoSQL 客户端
|
||
- **[microsoft/azure-eventhub-rust](https://officialskills.sh/microsoft/skills/azure-eventhub-rust)** - Event Hubs 流式客户端
|
||
- **[microsoft/azure-identity-rust](https://officialskills.sh/microsoft/skills/azure-identity-rust)** - Microsoft Entra ID 身份验证
|
||
- **[microsoft/azure-keyvault-certificates-rust](https://officialskills.sh/microsoft/skills/azure-keyvault-certificates-rust)** - Key Vault 证书管理
|
||
- **[microsoft/azure-keyvault-keys-rust](https://officialskills.sh/microsoft/skills/azure-keyvault-keys-rust)** - Key Vault 加密密钥管理
|
||
- **[microsoft/azure-keyvault-secrets-rust](https://officialskills.sh/microsoft/skills/azure-keyvault-secrets-rust)** - Key Vault 机密存储
|
||
- **[microsoft/azure-storage-blob-rust](https://officialskills.sh/microsoft/skills/azure-storage-blob-rust)** - Blob 对象存储客户端
|
||
|
||
### TypeScript 技能
|
||
|
||
- **[microsoft/azure-ai-contentsafety-ts](https://officialskills.sh/microsoft/skills/azure-ai-contentsafety-ts)** - 文本与图像内容安全
|
||
- **[microsoft/azure-ai-document-intelligence-ts](https://officialskills.sh/microsoft/skills/azure-ai-document-intelligence-ts)** - 文档文本与表格提取
|
||
- **[microsoft/azure-ai-projects-ts](https://officialskills.sh/microsoft/skills/azure-ai-projects-ts)** - AI Foundry 项目客户端与智能体
|
||
- **[microsoft/azure-ai-translation-ts](https://officialskills.sh/microsoft/skills/azure-ai-translation-ts)** - 文本与文档翻译
|
||
- **[microsoft/azure-ai-voicelive-ts](https://officialskills.sh/microsoft/skills/azure-ai-voicelive-ts)** - 实时双向语音 AI
|
||
- **[microsoft/azure-appconfiguration-ts](https://officialskills.sh/microsoft/skills/azure-appconfiguration-ts)** - 应用配置、功能开关、动态刷新
|
||
- **[microsoft/azure-cosmos-ts](https://officialskills.sh/microsoft/skills/azure-cosmos-ts)** - Cosmos DB NoSQL CRUD 与查询
|
||
- **[microsoft/azure-eventhub-ts](https://officialskills.sh/microsoft/skills/azure-eventhub-ts)** - 高吞吐量事件流
|
||
- **[microsoft/azure-identity-ts](https://officialskills.sh/microsoft/skills/azure-identity-ts)** - Microsoft Entra ID 身份验证
|
||
- **[microsoft/azure-keyvault-keys-ts](https://officialskills.sh/microsoft/skills/azure-keyvault-keys-ts)** - 加密密钥管理
|
||
- **[microsoft/azure-keyvault-secrets-ts](https://officialskills.sh/microsoft/skills/azure-keyvault-secrets-ts)** - 机密存储与检索
|
||
- **[microsoft/azure-microsoft-playwright-testing-ts](https://officialskills.sh/microsoft/skills/azure-microsoft-playwright-testing-ts)** - 在 Azure 上大规模运行 Playwright 测试
|
||
- **[microsoft/azure-monitor-opentelemetry-ts](https://officialskills.sh/microsoft/skills/azure-monitor-opentelemetry-ts)** - Application Insights 追踪与指标
|
||
- **[microsoft/azure-postgres-ts](https://officialskills.sh/microsoft/skills/azure-postgres-ts)** - PostgreSQL Flexible Server 连接
|
||
- **[microsoft/azure-search-documents-ts](https://officialskills.sh/microsoft/skills/azure-search-documents-ts)** - 带语义排序的向量/混合搜索
|
||
- **[microsoft/azure-servicebus-ts](https://officialskills.sh/microsoft/skills/azure-servicebus-ts)** - 队列与主题消息传递
|
||
- **[microsoft/azure-storage-blob-ts](https://officialskills.sh/microsoft/skills/azure-storage-blob-ts)** - Blob 上传、下载与管理
|
||
- **[microsoft/azure-storage-file-share-ts](https://officialskills.sh/microsoft/skills/azure-storage-file-share-ts)** - SMB 文件共享操作
|
||
- **[microsoft/azure-storage-queue-ts](https://officialskills.sh/microsoft/skills/azure-storage-queue-ts)** - 队列消息操作
|
||
- **[microsoft/azure-web-pubsub-ts](https://officialskills.sh/microsoft/skills/azure-web-pubsub-ts)** - 实时 WebSocket 发布/订阅消息
|
||
- **[microsoft/frontend-ui-dark-ts](https://officialskills.sh/microsoft/skills/frontend-ui-dark-ts)** - 深色主题 React,配合 Tailwind 与动画
|
||
- **[microsoft/m365-agents-ts](https://officialskills.sh/microsoft/skills/m365-agents-ts)** - M365、Teams 与 Copilot Studio 智能体
|
||
- **[microsoft/react-flow-node-ts](https://officialskills.sh/microsoft/skills/react-flow-node-ts)** - 配合 Zustand 的 React Flow 节点组件
|
||
- **[microsoft/zustand-store-ts](https://officialskills.sh/microsoft/skills/zustand-store-ts)** - 带中间件模式的 Zustand 存储
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">fal.ai 团队的 Skills</h3></summary>
|
||
|
||
- **[fal-ai-community/fal-3d](https://officialskills.sh/fal-ai-community/skills/fal-3d)** - 根据文本或图像生成 3D 模型
|
||
- **[fal-ai-community/fal-audio](https://officialskills.sh/fal-ai-community/skills/fal-audio)** - 使用 fal.ai 音频模型实现文本转语音(TTS)和语音转文本(STT)
|
||
- **[fal-ai-community/fal-generate](https://officialskills.sh/fal-ai-community/skills/fal-generate)** - 使用 fal.ai AI 模型生成图像和视频
|
||
- **[fal-ai-community/fal-image-edit](https://officialskills.sh/fal-ai-community/skills/fal-image-edit)** - AI 驱动的图像编辑,支持风格迁移和物体移除
|
||
- **[fal-ai-community/fal-kling-o3](https://officialskills.sh/fal-ai-community/skills/fal-kling-o3)** - 使用 Kling O3 生成图像和视频——Kling 最强大的模型系列
|
||
- **[fal-ai-community/fal-lip-sync](https://officialskills.sh/fal-ai-community/skills/fal-lip-sync)** - 创建说话头像视频,并将音频与视频进行唇形同步
|
||
- **[fal-ai-community/fal-platform](https://officialskills.sh/fal-ai-community/skills/fal-platform)** - 用于模型管理、定价和使用追踪的平台 API
|
||
- **[fal-ai-community/fal-realtime](https://officialskills.sh/fal-ai-community/skills/fal-realtime)** - 实时与流式 AI 图像生成
|
||
- **[fal-ai-community/fal-restore](https://officialskills.sh/fal-ai-community/skills/fal-restore)** - 修复和提升图像质量——去模糊、降噪、修复人脸、还原文档
|
||
- **[fal-ai-community/fal-train](https://officialskills.sh/fal-ai-community/skills/fal-train)** - 在 fal.ai 上训练自定义 AI 模型(LoRA)以实现个性化图像生成
|
||
- **[fal-ai-community/fal-tryon](https://officialskills.sh/fal-ai-community/skills/fal-tryon)** - 虚拟试穿——查看服装上身效果
|
||
- **[fal-ai-community/fal-upscale](https://officialskills.sh/fal-ai-community/skills/fal-upscale)** - 使用 AI 放大并增强图像和视频分辨率
|
||
- **[fal-ai-community/fal-video-edit](https://officialskills.sh/fal-ai-community/skills/fal-video-edit)** - 使用 AI 编辑现有视频——重混风格、放大、移除背景、添加音频
|
||
- **[fal-ai-community/fal-vision](https://officialskills.sh/fal-ai-community/skills/fal-vision)** - 分析图像——分割物体、检测、OCR、描述、视觉问答
|
||
- **[fal-ai-community/fal-workflow](https://officialskills.sh/fal-ai-community/skills/fal-workflow)** - 生成用于串联 AI 模型的工作流 JSON 文件
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">WordPress 开发团队的 Skills</h3></summary>
|
||
|
||
- **[WordPress/wordpress-router](https://officialskills.sh/WordPress/skills/wordpress-router)** - 分类 WordPress 仓库并路由到合适的工作流
|
||
- **[WordPress/wp-project-triage](https://officialskills.sh/WordPress/skills/wp-project-triage)** - 自动检测项目类型、工具链和版本
|
||
- **[WordPress/wp-block-development](https://officialskills.sh/WordPress/skills/wp-block-development)** - Gutenberg 区块:block.json、属性、渲染、弃用
|
||
- **[WordPress/wp-block-themes](https://officialskills.sh/WordPress/skills/wp-block-themes)** - 区块主题:theme.json、模板、样板、样式变体
|
||
- **[WordPress/wp-plugin-development](https://officialskills.sh/WordPress/skills/wp-plugin-development)** - 插件架构、钩子、设置 API、安全性
|
||
- **[WordPress/wp-rest-api](https://officialskills.sh/WordPress/skills/wp-rest-api)** - REST API 路由/端点、模式、认证和响应塑形
|
||
- **[WordPress/wp-interactivity-api](https://officialskills.sh/WordPress/skills/wp-interactivity-api)** - 使用 data-wp-* 指令和 store 实现前端交互
|
||
- **[WordPress/wp-abilities-api](https://officialskills.sh/WordPress/skills/wp-abilities-api)** - 基于能力的权限和 REST API 认证
|
||
- **[WordPress/wp-wpcli-and-ops](https://officialskills.sh/WordPress/skills/wp-wpcli-and-ops)** - WP-CLI 命令、自动化、多站点、搜索替换
|
||
- **[WordPress/wp-performance](https://officialskills.sh/WordPress/skills/wp-performance)** - 性能分析、缓存、数据库优化、Server-Timing
|
||
- **[WordPress/wp-phpstan](https://officialskills.sh/WordPress/skills/wp-phpstan)** - 面向 WordPress 项目的 PHPStan 静态分析
|
||
- **[WordPress/wp-playground](https://officialskills.sh/WordPress/skills/wp-playground)** - WordPress Playground,用于即时本地环境
|
||
- **[WordPress/wpds](https://officialskills.sh/WordPress/skills/wpds)** - WordPress 设计系统
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">OpenAI 的 Skills</h3></summary>
|
||
|
||
来自 OpenAI skills 仓库的官方精选技能。
|
||
|
||
- **[openai/cloudflare-deploy](https://officialskills.sh/openai/skills/cloudflare-deploy)** - 使用 Workers、Pages 和平台服务将应用部署到 Cloudflare
|
||
- **[openai/develop-web-game](https://officialskills.sh/openai/skills/develop-web-game)** - 使用 Playwright 配合时间步进迭代构建和测试网页游戏
|
||
- **[openai/doc](https://officialskills.sh/openai/skills/doc)** - 读取、创建和编辑 .docx 文档,保持格式与版式保真
|
||
- **[openai/gh-address-comments](https://officialskills.sh/openai/skills/gh-address-comments)** - 通过 CLI 处理开放 GitHub PR 上的评审和问题评论
|
||
- **[openai/gh-fix-ci](https://officialskills.sh/openai/skills/gh-fix-ci)** - 通过日志检查调试并修复失败的 GitHub Actions PR 检查
|
||
- **[openai/imagegen](https://officialskills.sh/openai/skills/imagegen)** - 使用 OpenAI Image API 为项目生成和编辑图像
|
||
- **[openai/jupyter-notebook](https://officialskills.sh/openai/skills/jupyter-notebook)** - 创建简洁、可复现的 Jupyter 笔记本,用于实验和教程
|
||
- **[openai/linear](https://officialskills.sh/openai/skills/linear)** - 在 Linear 中管理议题、项目和团队工作流
|
||
- **[openai/netlify-deploy](https://officialskills.sh/openai/skills/netlify-deploy)** - 通过 CLI 认证、关联和环境支持自动化 Netlify 部署
|
||
- **[openai/notion-knowledge-capture](https://officialskills.sh/openai/skills/notion-knowledge-capture)** - 将对话转换为结构化、可搜索的 Notion wiki 条目
|
||
- **[openai/notion-meeting-intelligence](https://officialskills.sh/openai/skills/notion-meeting-intelligence)** - 拉取 Notion 上下文并定制议程以准备会议
|
||
- **[openai/notion-research-documentation](https://officialskills.sh/openai/skills/notion-research-documentation)** - 研究 Notion 内容并将发现综合为结构化简报
|
||
- **[openai/notion-spec-to-implementation](https://officialskills.sh/openai/skills/notion-spec-to-implementation)** - 将 Notion 规格说明转换为关联的实现计划和任务
|
||
- **[openai/openai-docs](https://officialskills.sh/openai/skills/openai-docs)** - 提供来自 OpenAI 开发者文档的权威指导
|
||
- **[openai/pdf](https://officialskills.sh/openai/skills/pdf)** - 读取、创建和审阅 PDF,保持版式与视觉格式完整性
|
||
- **[openai/playwright](https://officialskills.sh/openai/skills/playwright)** - 自动化真实浏览器交互,用于导航、表单和抓取
|
||
- **[openai/render-deploy](https://officialskills.sh/openai/skills/render-deploy)** - 使用 Git 支持的服务将应用部署到 Render 云平台
|
||
- **[openai/screenshot](https://officialskills.sh/openai/skills/screenshot)** - 跨操作系统平台截取桌面、应用窗口或像素区域
|
||
- **[openai/security-best-practices](https://officialskills.sh/openai/skills/security-best-practices)** - 审查代码中的语言特定安全漏洞
|
||
- **[openai/security-ownership-map](https://officialskills.sh/openai/skills/security-ownership-map)** - 映射人员与文件归属关系,计算巴士因子并识别风险
|
||
- **[openai/security-threat-model](https://officialskills.sh/openai/skills/security-threat-model)** - 生成仓库特定的威胁模型,识别信任边界
|
||
- **[openai/sentry](https://officialskills.sh/openai/skills/sentry)** - 检查 Sentry 议题,汇总生产错误并拉取健康数据
|
||
- **[openai/sora](https://officialskills.sh/openai/skills/sora)** - 通过 OpenAI Sora API 生成、重混和管理短视频片段
|
||
- **[openai/speech](https://officialskills.sh/openai/skills/speech)** - 使用 OpenAI API 及内置语音从文本生成语音音频
|
||
- **[openai/spreadsheet](https://officialskills.sh/openai/skills/spreadsheet)** - 使用公式创建、编辑、分析和可视化电子表格
|
||
- **[openai/transcribe](https://officialskills.sh/openai/skills/transcribe)** - 将音频文件转写为文本,可选说话人分离
|
||
- **[openai/vercel-deploy](https://officialskills.sh/openai/skills/vercel-deploy)** - 将应用和网站部署到 Vercel,支持预览或生产选项
|
||
- **[openai/yeet](https://officialskills.sh/openai/skills/yeet)** - 通过 CLI 暂存、提交、推送代码并打开 GitHub 拉取请求
|
||
- **[openai/aspnet-core](https://officialskills.sh/openai/skills/aspnet-core)** - 构建、审查和架构 ASP.NET Core 应用(Blazor、MVC、Minimal APIs 等)
|
||
- **[openai/chatgpt-apps](https://officialskills.sh/openai/skills/chatgpt-apps)** - 使用 MCP 服务器和 widget UI 构建、脚手架和排查 ChatGPT Apps SDK 应用
|
||
- **[openai/figma](https://officialskills.sh/openai/skills/figma)** - 使用 Figma MCP 服务器获取设计上下文并将节点转换为生产代码
|
||
- **[openai/figma-code-connect-components](https://officialskills.sh/openai/skills/figma-code-connect-components)** - 使用 Code Connect 将 Figma 设计组件连接到代码组件
|
||
- **[openai/figma-create-design-system-rules](https://officialskills.sh/openai/skills/figma-create-design-system-rules)** - 使用 Figma MCP 服务器实现 Figma 设计的规则
|
||
- **[openai/figma-create-new-file](https://officialskills.sh/openai/skills/figma-create-new-file)** - 创建新的空白 Figma 文件或 FigJam 文件
|
||
- **[openai/figma-generate-design](https://officialskills.sh/openai/skills/figma-generate-design)** - 使用设计系统 token 将应用页面和布局转换为 Figma
|
||
- **[openai/figma-generate-library](https://officialskills.sh/openai/skills/figma-generate-library)** - 从代码库在 Figma 中构建或更新专业级设计系统
|
||
- **[openai/figma-implement-design](https://officialskills.sh/openai/skills/figma-implement-design)** - 将 Figma 设计转换为生产就绪代码,实现 1:1 视觉保真
|
||
- **[openai/figma-use](https://officialskills.sh/openai/skills/figma-use)** - 每次 use_figma 工具调用的前置技能——在 Figma 上下文中执行写入/读取操作
|
||
- **[openai/frontend-skill](https://officialskills.sh/openai/skills/frontend-skill)** - 以克制的构图创建视觉出色的落地页、网站和应用 UI
|
||
- **[openai/playwright-interactive](https://officialskills.sh/openai/skills/playwright-interactive)** - 通过 js_repl 实现持久化浏览器和 Electron 交互,用于迭代式 UI 调试
|
||
- **[openai/slides](https://officialskills.sh/openai/skills/slides)** - 使用 PptxGenJS 创建和编辑 .pptx 演示文稿
|
||
- **[openai/winui-app](https://officialskills.sh/openai/skills/winui-app)** - 使用 C# 和 Windows App SDK 引导并开发现代 WinUI 3 桌面应用
|
||
</details>
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Figma 技能</h3></summary>
|
||
|
||
Figma MCP 服务器指南中的官方技能。
|
||
|
||
- **[figma/figma-code-connect-components](https://officialskills.sh/figma/skills/figma-code-connect-components)** - 使用 Code Connect 将 Figma 设计组件与代码组件关联
|
||
- **[figma/figma-create-design-system-rules](https://officialskills.sh/figma/skills/figma-create-design-system-rules)** - 为 Figma 转代码工作流生成项目专属的设计系统规则
|
||
- **[figma/figma-create-new-file](https://officialskills.sh/figma/skills/figma-create-new-file)** - 创建新的空白 Figma Design 或 FigJam 文件
|
||
- **[figma/figma-generate-design](https://officialskills.sh/figma/skills/figma-generate-design)** - 基于代码或描述,使用设计系统组件在 Figma 中构建或更新界面
|
||
- **[figma/figma-generate-library](https://officialskills.sh/figma/skills/figma-generate-library)** - 从代码库在 Figma 中构建或更新设计系统库
|
||
- **[figma/figma-implement-design](https://officialskills.sh/figma/skills/figma-implement-design)** - 将 Figma 设计高保真(1:1)翻译为可用于生产的应用代码
|
||
- **[figma/figma-use](https://officialskills.sh/figma/skills/figma-use)** - 运行 Figma Plugin API 脚本,完成画布写入、检查、变量及设计系统相关操作
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Corey Haines 的营销技能</h3></summary>
|
||
|
||
[Corey Haines](https://github.com/coreyhaines31), 的官方营销技能,覆盖完整 SaaS 营销栈:从 SEO 与文案写作到增长、CRO(转化率优化)和付费获客。
|
||
|
||
- **[coreyhaines31/ab-test-setup](https://github.com/coreyhaines31/marketingskills/tree/main/skills/ab-test-setup)** - 为任意数字化体验规划并实施 A/B 测试或实验
|
||
- **[coreyhaines31/ad-creative](https://github.com/coreyhaines31/marketingskills/tree/main/skills/ad-creative)** - 生成并迭代广告创意,包括标题、描述和正文
|
||
- **[coreyhaines31/ai-seo](https://github.com/coreyhaines31/marketingskills/tree/main/skills/ai-seo)** - 优化内容以出现在 AI 生成答案和 LLM 搜索结果中
|
||
- **[coreyhaines31/analytics-tracking](https://github.com/coreyhaines31/marketingskills/tree/main/skills/analytics-tracking)** - 搭建并审计分析追踪与度量流水线
|
||
- **[coreyhaines31/churn-prevention](https://github.com/coreyhaines31/marketingskills/tree/main/skills/churn-prevention)** - 构建取消流程、挽留优惠并挽回失败付款
|
||
- **[coreyhaines31/cold-email](https://github.com/coreyhaines31/marketingskills/tree/main/skills/cold-email)** - 撰写能促成转化的 B2B 冷邮件及跟进序列
|
||
- **[coreyhaines31/competitor-alternatives](https://github.com/coreyhaines31/marketingskills/tree/main/skills/competitor-alternatives)** - 为 SEO 构建竞品对比与替代方案落地页
|
||
- **[coreyhaines31/content-strategy](https://github.com/coreyhaines31/marketingskills/tree/main/skills/content-strategy)** - 规划内容策略,决定优先创作的主题与形式
|
||
- **[coreyhaines31/copy-editing](https://github.com/coreyhaines31/marketingskills/tree/main/skills/copy-editing)** - 编辑并改进现有营销文案,提升清晰度与感染力
|
||
- **[coreyhaines31/copywriting](https://github.com/coreyhaines31/marketingskills/tree/main/skills/copywriting)** - 为落地页、首页和广告撰写或改写营销文案
|
||
- **[coreyhaines31/email-sequence](https://github.com/coreyhaines31/marketingskills/tree/main/skills/email-sequence)** - 构建邮件序列、滴灌营销活动和生命周期邮件流程
|
||
- **[coreyhaines31/form-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/form-cro)** - 优化线索捕获与联系表单以提升转化率
|
||
- **[coreyhaines31/free-tool-strategy](https://github.com/coreyhaines31/marketingskills/tree/main/skills/free-tool-strategy)** - 规划并构建用于获客与 SEO 价值的免费工具
|
||
- **[coreyhaines31/launch-strategy](https://github.com/coreyhaines31/marketingskills/tree/main/skills/launch-strategy)** - 规划产品发布、功能公告及上市(go-to-market)策略
|
||
- **[coreyhaines31/marketing-ideas](https://github.com/coreyhaines31/marketingskills/tree/main/skills/marketing-ideas)** - 为 SaaS 产品生成营销策略与活动创意
|
||
- **[coreyhaines31/marketing-psychology](https://github.com/coreyhaines31/marketingskills/tree/main/skills/marketing-psychology)** - 将心理学原理与行为科学应用于文案与设计
|
||
- **[coreyhaines31/onboarding-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/onboarding-cro)** - 优化注册后引导与用户激活,缩短价值实现时间(time-to-value)
|
||
- **[coreyhaines31/page-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/page-cro)** - 提升各类营销页面(含首页与落地页)的转化率
|
||
- **[coreyhaines31/paid-ads](https://github.com/coreyhaines31/marketingskills/tree/main/skills/paid-ads)** - 在 Google、Meta、LinkedIn 等平台创建并优化付费广告活动
|
||
- **[coreyhaines31/paywall-upgrade-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/paywall-upgrade-cro)** - 设计并优化升级界面、付费墙与追加销售弹窗
|
||
- **[coreyhaines31/popup-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/popup-cro)** - 创建并优化用于转化的弹窗、模态框与滑入式组件
|
||
- **[coreyhaines31/pricing-strategy](https://github.com/coreyhaines31/marketingskills/tree/main/skills/pricing-strategy)** - 为 SaaS 产品定义定价、打包与变现策略
|
||
- **[coreyhaines31/product-marketing-context](https://github.com/coreyhaines31/marketingskills/tree/main/skills/product-marketing-context)** - 创建并维护产品营销背景文档,确保信息传达一致
|
||
- **[coreyhaines31/programmatic-seo](https://github.com/coreyhaines31/marketingskills/tree/main/skills/programmatic-seo)** - 构建 SEO 驱动的页面模板,用于大规模内容生成
|
||
- **[coreyhaines31/referral-program](https://github.com/coreyhaines31/marketingskills/tree/main/skills/referral-program)** - 设计并优化推荐、联盟与口碑传播计划
|
||
- **[coreyhaines31/revops](https://github.com/coreyhaines31/marketingskills/tree/main/skills/revops)** - 精简收入运营、线索生命周期及市场到销售的交接流程
|
||
- **[coreyhaines31/sales-enablement](https://github.com/coreyhaines31/marketingskills/tree/main/skills/sales-enablement)** - 创建路演材料、单页简介、异议处理文档与演示脚本
|
||
- **[coreyhaines31/schema-markup](https://github.com/coreyhaines31/marketingskills/tree/main/skills/schema-markup)** - 添加并优化 schema 标记与结构化数据以改善 SEO
|
||
- **[coreyhaines31/seo-audit](https://github.com/coreyhaines31/marketingskills/tree/main/skills/seo-audit)** - 审计并诊断网站的技术 SEO 与页面内 SEO 问题
|
||
- **[coreyhaines31/signup-flow-cro](https://github.com/coreyhaines31/marketingskills/tree/main/skills/signup-flow-cro)** - 优化注册、登记与试用激活流程以提高转化率
|
||
- **[coreyhaines31/site-architecture](https://github.com/coreyhaines31/marketingskills/tree/main/skills/site-architecture)** - 规划并重构页面层级、导航与 URL 结构
|
||
- **[coreyhaines31/social-content](https://github.com/coreyhaines31/marketingskills/tree/main/skills/social-content)** - 为 LinkedIn、Twitter/X 和 Instagram 创建并排期社交媒体内容
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Kim Barrett 的广告技能</h3></summary>
|
||
|
||
Kim Barrett 的直接响应式广告技能,按 foundations、copy-chief、operator-os、orchestrators 和 QA 组织——涵盖用户画像(avatar)工作、offer 设计、Schwartz 风格文案、创意测试及全漏斗活动编排。
|
||
|
||
- **[realkimbarrett/avatar-extraction](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/foundations/avatar-extraction)** - 精确定义买家是谁、他们想要什么、尝试过什么,以及驱动其决策的因素
|
||
- **[realkimbarrett/offer-extraction](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/foundations/offer-extraction)** - 将产品或服务打造成有说服力、高转化的 offer(报价/要约)
|
||
- **[realkimbarrett/schwartz-awareness-mapper](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/copy-chief/schwartz-awareness-mapper)** - 确定受众认知层级(awareness level)及正确的信息传递方式
|
||
- **[realkimbarrett/mechanism-builder](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/copy-chief/mechanism-builder)** - 用独特的 mechanism(机制论述)解释为何你的方案有效而他人失败
|
||
- **[realkimbarrett/headline-matrix](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/copy-chief/headline-matrix)** - 从不同角度生成高表现的标题变体
|
||
- **[realkimbarrett/objection-crusher](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/copy-chief/objection-crusher)** - 识别并化解买家异议与犹豫
|
||
- **[realkimbarrett/ad-angle-multiplier](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/operator-os/ad-angle-multiplier)** - 将核心创意扩展为多个不同广告角度,用于创意测试
|
||
- **[realkimbarrett/scroll-stopping-creative](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/operator-os/scroll-stopping-creative)** - 打造能在前 3 秒抓住注意力的广告概念
|
||
- **[realkimbarrett/conversion-path-builder](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/operator-os/conversion-path-builder)** - 设计从点击到转化与预约通话的最优漏斗
|
||
- **[realkimbarrett/performance-diagnosis](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/operator-os/performance-diagnosis)** - 诊断营销活动表现不佳的原因——低转化、高 CPL(单次潜客成本)、广告效果差
|
||
- **[realkimbarrett/full-funnel-campaign-orchestrator](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/orchestrators/full-funnel-campaign-orchestrator)** - 协调所有技能,端到端构建完整的广告 + 漏斗营销活动
|
||
- **[realkimbarrett/generic-language-killer](https://github.com/realkimbarrett/advertising-skills/tree/main/skills/qa/generic-language-killer)** - 去除模糊、企业腔或 AI 腔表述,替换为清晰、具体、有人味的措辞
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by Binance</h3></summary>
|
||
|
||
Binance 团队提供的官方 Web3 与交易技能。涵盖加密货币市场数据、链上分析、代币安全审计,以及通过 Binance API 进行的现货交易。
|
||
|
||
- **[binance/crypto-market-rank](https://officialskills.sh/binance/skills/crypto-market-rank)** - 查询加密货币市场排名,包括热门代币、聪明钱流入、meme 币排名及顶级交易者 PnL 排行榜
|
||
- **[binance/meme-rush](https://officialskills.sh/binance/skills/meme-rush)** - 追踪来自 launchpad(Pump.fun、Four.meme)的实时 meme 代币列表,以及按净流入排名的 AI 驱动热门市场主题
|
||
- **[binance/query-address-info](https://officialskills.sh/binance/skills/query-address-info)** - 查询 BSC、Base 或 Solana 上任意钱包地址的全部代币持仓与投资组合头寸
|
||
- **[binance/query-token-audit](https://officialskills.sh/binance/skills/query-token-audit)** - 审计代币安全性,在 BSC、Base、Solana 和 Ethereum 上检测骗局、蜜罐和恶意合约
|
||
- **[binance/query-token-info](https://officialskills.sh/binance/skills/query-token-info)** - 通过关键词或合约地址搜索代币,获取元数据、实时市场数据和 K 线蜡烛图
|
||
- **[binance/trading-signal](https://officialskills.sh/binance/skills/trading-signal)** - 监控链上 Smart Money 买卖信号,包含 Solana 和 BSC 上的价格、最大涨幅与退出率数据
|
||
- **[binance/spot](https://officialskills.sh/binance/skills/spot)** - 通过 API key 认证在 Binance 下单并管理现货交易订单,支持主网与测试网
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by Apollo GraphQL</h3></summary>
|
||
|
||
Apollo GraphQL 团队的官方技能,用于构建 GraphQL 客户端、服务端、联邦 supergraph 以及 Apollo Router。
|
||
|
||
- **[apollographql/apollo-client](https://officialskills.sh/apollographql/skills/apollo-client)** - 使用 Apollo Client 4 构建 React 应用
|
||
- **[apollographql/apollo-connectors](https://officialskills.sh/apollographql/skills/apollo-connectors)** - 使用 Apollo Connectors 将 REST API 集成到 GraphQL supergraph
|
||
- **[apollographql/apollo-federation](https://officialskills.sh/apollographql/skills/apollo-federation)** - 编写 Apollo Federation 2 子图 schema,并将其组合为 supergraph
|
||
- **[apollographql/apollo-kotlin](https://officialskills.sh/apollographql/skills/apollo-kotlin)** - 面向 Android、JVM 和 Kotlin Multiplatform 项目的 GraphQL 客户端
|
||
- **[apollographql/apollo-mcp-server](https://officialskills.sh/apollographql/skills/apollo-mcp-server)** - 通过 Model Context Protocol(MCP)将 AI 智能体连接到 GraphQL API
|
||
- **[apollographql/apollo-router](https://officialskills.sh/apollographql/skills/apollo-router)** - 面向基于 Rust 的 Apollo Router 的版本感知配置生成器
|
||
- **[apollographql/apollo-router-plugin-creator](https://officialskills.sh/apollographql/skills/apollo-router-plugin-creator)** - 为 Apollo Router 编写原生 Rust 插件
|
||
- **[apollographql/apollo-server](https://officialskills.sh/apollographql/skills/apollo-server)** - 使用 Apollo Server 5 构建 GraphQL 服务端
|
||
- **[apollographql/graphql-operations](https://officialskills.sh/apollographql/skills/graphql-operations)** - 遵循最佳实践编写 GraphQL 查询、变更与订阅
|
||
- **[apollographql/graphql-schema](https://officialskills.sh/apollographql/skills/graphql-schema)** - 设计清晰、可演进的 GraphQL schema 的参考指南
|
||
- **[apollographql/rover](https://officialskills.sh/apollographql/skills/rover)** - 在 Apollo GraphOS 中管理 GraphQL schema 的 CLI 工具
|
||
- **[apollographql/rust-best-practices](https://officialskills.sh/apollographql/skills/rust-best-practices)** - 摘自 Apollo GraphQL 内部手册的 Rust 编码指南
|
||
- **[apollographql/skill-creator](https://officialskills.sh/apollographql/skills/skill-creator)** - 创建并结构化面向 Apollo GraphQL 的 Agent Skills
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by Auth0</h3></summary>
|
||
|
||
Auth0 团队的官方认证与身份技能。涵盖主流框架 SDK,以及 MFA、迁移与快速入门检测等工作流。
|
||
|
||
- **[auth0/auth0-android](https://officialskills.sh/auth0/skills/auth0-android)** - 使用 Auth0 SDK 为原生 Android 应用添加认证
|
||
- **[auth0/auth0-angular](https://officialskills.sh/auth0/skills/auth0-angular)** - 使用 @auth0/auth0-angular 为 Angular 应用添加认证
|
||
- **[auth0/auth0-aspnetcore-api](https://officialskills.sh/auth0/skills/auth0-aspnetcore-api)** - 为 ASP.NET Core API 添加 JWT 访问令牌校验
|
||
- **[auth0/auth0-express](https://officialskills.sh/auth0/skills/auth0-express)** - 为 Express.js 应用添加基于 session 的认证
|
||
- **[auth0/auth0-fastify](https://officialskills.sh/auth0/skills/auth0-fastify)** - 为 Fastify Web 应用添加基于 session 的认证
|
||
- **[auth0/auth0-fastify-api](https://officialskills.sh/auth0/skills/auth0-fastify-api)** - 使用 JWT Bearer 令牌校验保护 Fastify API 端点
|
||
- **[auth0/auth0-mfa](https://officialskills.sh/auth0/skills/auth0-mfa)** - 为基于 Auth0 的应用添加多因素认证(MFA)
|
||
- **[auth0/auth0-migration](https://officialskills.sh/auth0/skills/auth0-migration)** - 将用户与认证流程从其他提供商迁移到 Auth0
|
||
- **[auth0/auth0-nextjs](https://officialskills.sh/auth0/skills/auth0-nextjs)** - 为 Next.js 应用添加认证
|
||
- **[auth0/auth0-nuxt](https://officialskills.sh/auth0/skills/auth0-nuxt)** - 为 Nuxt 3/4 应用添加 Auth0 认证,支持加密 cookie session
|
||
- **[auth0/auth0-quickstart](https://officialskills.sh/auth0/skills/auth0-quickstart)** - 自动检测你的框架并搭建 Auth0 集成脚手架
|
||
- **[auth0/auth0-react](https://officialskills.sh/auth0/skills/auth0-react)** - 使用 @auth0/auth0-react 为 React SPA 添加认证
|
||
- **[auth0/auth0-react-native](https://officialskills.sh/auth0/skills/auth0-react-native)** - 为 React Native 与 Expo 移动应用添加认证
|
||
- **[auth0/auth0-vue](https://officialskills.sh/auth0/skills/auth0-vue)** - 为 Vue.js 应用添加认证
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Brave 出品的 Skills</h3></summary>
|
||
|
||
Brave 团队官方 Skills,用于访问 Brave Search API,涵盖网页、图片、视频、新闻以及本地兴趣点(POI)数据。
|
||
|
||
- **[brave/answers](https://officialskills.sh/brave/skills/answers)** - 基于实时网页搜索结果的 AI 生成答案
|
||
- **[brave/bx](https://officialskills.sh/brave/skills/bx)** - 面向 AI 智能体构建的网页搜索 CLI 工具
|
||
- **[brave/images-search](https://officialskills.sh/brave/skills/images-search)** - 使用 Brave Search API 搜索图片
|
||
- **[brave/llm-context](https://officialskills.sh/brave/skills/llm-context)** - 从 Brave Search 返回预提取的网页内容(文本、表格、代码)
|
||
- **[brave/local-descriptions](https://officialskills.sh/brave/skills/local-descriptions)** - 获取兴趣点的 AI 生成文本描述
|
||
- **[brave/local-pois](https://officialskills.sh/brave/skills/local-pois)** - 检索详细的本地商户与 POI 信息
|
||
- **[brave/news-search](https://officialskills.sh/brave/skills/news-search)** - 搜索 Brave 新闻索引并获取文章元数据
|
||
- **[brave/spellcheck](https://officialskills.sh/brave/skills/spellcheck)** - 检查搜索查询的拼写错误并获取纠正建议
|
||
- **[brave/suggest](https://officialskills.sh/brave/skills/suggest)** - 通过 Brave Search API 查询自动补全建议
|
||
- **[brave/videos-search](https://officialskills.sh/brave/skills/videos-search)** - 通过 Brave Search API 搜索全网视频
|
||
- **[brave/web-search](https://officialskills.sh/brave/skills/web-search)** - 通过 Brave Search API 搜索网页并返回排序结果
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Browserbase 出品的 Skills</h3></summary>
|
||
|
||
Browserbase 团队官方浏览器自动化 Skills,涵盖无头浏览、Cookie 同步、无服务器函数以及对抗式 UI 测试。
|
||
|
||
- **[browserbase/browser](https://officialskills.sh/browserbase/skills/browser)** - 通过自然语言 CLI 命令自动化网页浏览器交互
|
||
- **[browserbase/browserbase-cli](https://officialskills.sh/browserbase/skills/browserbase-cli)** - Browserbase 平台的 CLI 封装
|
||
- **[browserbase/cookie-sync](https://officialskills.sh/browserbase/skills/cookie-sync)** - 将本地 Chrome 的 Cookie 导出到 Browserbase 持久化上下文
|
||
- **[browserbase/fetch](https://officialskills.sh/browserbase/skills/fetch)** - 通过 Browserbase API 获取 HTML、JSON、响应头与状态码
|
||
- **[browserbase/functions](https://officialskills.sh/browserbase/skills/functions)** - 将浏览器自动化脚本部署为无服务器云函数
|
||
- **[browserbase/search](https://officialskills.sh/browserbase/skills/search)** - 通过 Browserbase API 搜索网页并返回结构化结果
|
||
- **[browserbase/ui-test](https://officialskills.sh/browserbase/skills/ui-test)** - 在真实浏览器中分析 git diff 以运行对抗式 UI 测试
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">CodeRabbit 出品的 Skills</h3></summary>
|
||
|
||
CodeRabbit 团队官方 AI 代码审查 Skills。
|
||
|
||
- **[coderabbitai/autofix](https://officialskills.sh/coderabbitai/skills/autofix)** - 从 GitHub PR 获取未解决的 CodeRabbit 审查评论并应用修复
|
||
- **[coderabbitai/code-review](https://officialskills.sh/coderabbitai/skills/code-review)** - 通过 CodeRabbit CLI 运行 AI 驱动的代码审查
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Coinbase 出品的 Skills</h3></summary>
|
||
|
||
Coinbase 团队官方钱包、支付与交易 Skills,涵盖 USDC 转账、链上查询、x402 付费 API 以及 Base 交易。
|
||
|
||
- **[coinbase/authenticate-wallet](https://officialskills.sh/coinbase/skills/authenticate-wallet)** - 通过邮箱 OTP 处理 Coinbase 支付钱包的登录
|
||
- **[coinbase/fund](https://officialskills.sh/coinbase/skills/fund)** - 通过 Coinbase Onramp 向 Coinbase 驱动的钱包充值 USDC
|
||
- **[coinbase/monetize-service](https://officialskills.sh/coinbase/skills/monetize-service)** - 搭建 Express 服务器,使用 x402 按请求收取 USDC
|
||
- **[coinbase/pay-for-service](https://officialskills.sh/coinbase/skills/pay-for-service)** - 调用采用 x402 协议且自动使用 USDC 的付费 API 端点
|
||
- **[coinbase/query-onchain-data](https://officialskills.sh/coinbase/skills/query-onchain-data)** - 在 Base 上查询已解码的链上数据(事件、交易、区块)
|
||
- **[coinbase/search-for-service](https://officialskills.sh/coinbase/skills/search-for-service)** - 搜索并浏览 x402 bazaar 市场
|
||
- **[coinbase/send-usdc](https://officialskills.sh/coinbase/skills/send-usdc)** - 向 Base 上任意以太坊地址或 ENS 名称发送 USDC
|
||
- **[coinbase/trade](https://officialskills.sh/coinbase/skills/trade)** - 使用 CDP Swap API 在 Base 上兑换与交易代币
|
||
- **[coinbase/x402](https://officialskills.sh/coinbase/skills/x402)** - 使用 x402 支付协议发现并调用付费 API 端点
|
||
|
||
</details>
|
||
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Datadog Labs 出品的 Skills</h3></summary>
|
||
|
||
Datadog Labs 可观测性 Skills,用于 APM、日志、监控器以及由 pup CLI 驱动的 LLM 可观测性工作流。
|
||
|
||
- **[datadog-labs/dd-apm](https://officialskills.sh/datadog-labs/skills/dd-apm)** - 在编辑器中直接查询 Datadog APM 数据
|
||
- **[datadog-labs/dd-docs](https://officialskills.sh/datadog-labs/skills/dd-docs)** - 通过面向 LLM 优化的文档索引查阅 Datadog 文档
|
||
- **[datadog-labs/dd-llmo-eval-bootstrap](https://officialskills.sh/datadog-labs/skills/dd-llmo-eval-bootstrap)** - 分析生产环境 LLM 追踪并生成评估器
|
||
- **[datadog-labs/dd-llmo-eval-trace-rca](https://officialskills.sh/datadog-labs/skills/dd-llmo-eval-trace-rca)** - 使用评估追踪对 LLM 应用故障进行根因分析
|
||
- **[datadog-labs/dd-llmo-experiment-analyzer](https://officialskills.sh/datadog-labs/skills/dd-llmo-experiment-analyzer)** - 分析单个或对比性的 LLM 实验结果
|
||
- **[datadog-labs/dd-logs](https://officialskills.sh/datadog-labs/skills/dd-logs)** - 通过 pup CLI 搜索、筛选并归档 Datadog 日志
|
||
- **[datadog-labs/dd-monitors](https://officialskills.sh/datadog-labs/skills/dd-monitors)** - 通过 pup CLI 管理 Datadog 监控器
|
||
- **[datadog-labs/dd-pup](https://officialskills.sh/datadog-labs/skills/dd-pup)** - 基于 Rust 的 CLI(pup),用于与 Datadog API 交互
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Firebase 出品的 Skills</h3></summary>
|
||
|
||
Firebase 团队官方 Skills,涵盖设置、身份验证、Firestore、托管、Genkit AI SDK 以及安全规则审计。
|
||
|
||
- **[firebase/developing-genkit-dart](https://officialskills.sh/firebase/skills/developing-genkit-dart)** - 使用 Genkit Dart SDK 构建 AI 应用
|
||
- **[firebase/developing-genkit-go](https://officialskills.sh/firebase/skills/developing-genkit-go)** - 使用 Genkit Go SDK 构建 AI 应用
|
||
- **[firebase/developing-genkit-js](https://officialskills.sh/firebase/skills/developing-genkit-js)** - 在 Node.js 中使用 Firebase Genkit 构建 AI 应用
|
||
- **[firebase/firebase-ai-logic-basics](https://officialskills.sh/firebase/skills/firebase-ai-logic-basics)** - 通过 Firebase AI Logic 从 Web 与移动应用调用 Gemini 模型
|
||
- **[firebase/firebase-app-hosting-basics](https://officialskills.sh/firebase/skills/firebase-app-hosting-basics)** - 部署并管理全栈 Web 应用(Next.js、Angular 等)
|
||
- **[firebase/firebase-auth-basics](https://officialskills.sh/firebase/skills/firebase-auth-basics)** - 配置 Firebase Authentication 及登录提供方
|
||
- **[firebase/firebase-basics](https://officialskills.sh/firebase/skills/firebase-basics)** - 处理 Firebase CLI 安装、身份验证与日常开发工作流
|
||
- **[firebase/firebase-data-connect-basics](https://officialskills.sh/firebase/skills/firebase-data-connect-basics)** - 构建由 Cloud SQL 支撑的 Firebase Data Connect 后端
|
||
- **[firebase/firebase-firestore-enterprise-native-mode](https://officialskills.sh/firebase/skills/firebase-firestore-enterprise-native-mode)** - 设置并使用 Firestore Enterprise Native Mode
|
||
- **[firebase/firebase-firestore-standard](https://officialskills.sh/firebase/skills/firebase-firestore-standard)** - Cloud Firestore Standard Edition 完整指南
|
||
- **[firebase/firebase-hosting-basics](https://officialskills.sh/firebase/skills/firebase-hosting-basics)** - 将静态站点、SPA 与微服务部署到 Firebase Hosting
|
||
- **[firebase/firebase-security-rules-auditor](https://officialskills.sh/firebase/skills/firebase-security-rules-auditor)** - 审计 Firestore 安全规则并标记高风险模式
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Flutter 技能</h3></summary>
|
||
|
||
Flutter 团队官方技能,涵盖跨平台 Flutter 应用的布局、状态、导航、原生互操作、平台配置与测试。
|
||
|
||
- **[flutter/flutter-adding-home-screen-widgets](https://officialskills.sh/flutter/skills/flutter-adding-home-screen-widgets)** - 为 Android 和 iOS 上的 Flutter 应用添加主屏幕小组件
|
||
- **[flutter/flutter-animating-apps](https://officialskills.sh/flutter/skills/flutter-animating-apps)** - 实现动画效果、过渡与动效
|
||
- **[flutter/flutter-architecting-apps](https://officialskills.sh/flutter/skills/flutter-architecting-apps)** - 使用分层架构组织 Flutter 应用结构
|
||
- **[flutter/flutter-building-forms](https://officialskills.sh/flutter/skills/flutter-building-forms)** - 构建带校验与用户输入的 Flutter 表单
|
||
- **[flutter/flutter-building-layouts](https://officialskills.sh/flutter/skills/flutter-building-layouts)** - 使用约束系统(Row、Column、Stack)构建并修复布局
|
||
- **[flutter/flutter-building-plugins](https://officialskills.sh/flutter/skills/flutter-building-plugins)** - 创建连接 Dart 与平台代码的 Flutter 插件
|
||
- **[flutter/flutter-caching-data](https://officialskills.sh/flutter/skills/flutter-caching-data)** - 实现离线优先(offline-first)的缓存策略
|
||
- **[flutter/flutter-embedding-native-views](https://officialskills.sh/flutter/skills/flutter-embedding-native-views)** - 在 Flutter 组件中嵌入原生 Android、iOS 与 macOS 视图
|
||
- **[flutter/flutter-handling-concurrency](https://officialskills.sh/flutter/skills/flutter-handling-concurrency)** - 在后台 Dart isolate 中运行繁重任务
|
||
- **[flutter/flutter-handling-http-and-json](https://officialskills.sh/flutter/skills/flutter-handling-http-and-json)** - 处理 HTTP 请求与 JSON 序列化
|
||
- **[flutter/flutter-implementing-navigation-and-routing](https://officialskills.sh/flutter/skills/flutter-implementing-navigation-and-routing)** - 处理路由、导航与深度链接(deep linking)
|
||
- **[flutter/flutter-improving-accessibility](https://officialskills.sh/flutter/skills/flutter-improving-accessibility)** - 为屏幕阅读器与辅助技术配置 Flutter
|
||
- **[flutter/flutter-interoperating-with-native-apis](https://officialskills.sh/flutter/skills/flutter-interoperating-with-native-apis)** - 将 Flutter 与原生平台 API 桥接
|
||
- **[flutter/flutter-localizing-apps](https://officialskills.sh/flutter/skills/flutter-localizing-apps)** - 为多种语言与地区配置 Flutter
|
||
- **[flutter/flutter-managing-state](https://officialskills.sh/flutter/skills/flutter-managing-state)** - 管理局部组件状态与共享应用状态
|
||
- **[flutter/flutter-reducing-app-size](https://officialskills.sh/flutter/skills/flutter-reducing-app-size)** - 测量并优化 Flutter 应用包体积
|
||
- **[flutter/flutter-setting-up-on-linux](https://officialskills.sh/flutter/skills/flutter-setting-up-on-linux)** - 配置 Linux 机器以进行 Flutter 桌面开发
|
||
- **[flutter/flutter-setting-up-on-macos](https://officialskills.sh/flutter/skills/flutter-setting-up-on-macos)** - 配置 macOS 机器以进行 Flutter 开发
|
||
- **[flutter/flutter-setting-up-on-windows](https://officialskills.sh/flutter/skills/flutter-setting-up-on-windows)** - 配置 Windows 机器以进行 Flutter 开发
|
||
- **[flutter/flutter-testing-apps](https://officialskills.sh/flutter/skills/flutter-testing-apps)** - 实现单元测试、组件测试与集成测试
|
||
- **[flutter/flutter-theming-apps](https://officialskills.sh/flutter/skills/flutter-theming-apps)** - 通过主题系统自定义 Flutter 应用外观
|
||
- **[flutter/flutter-working-with-databases](https://officialskills.sh/flutter/skills/flutter-working-with-databases)** - 使用 SQLite 构建结构化数据层
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Dean Peters 产品经理技能</h3></summary>
|
||
|
||
46 个经过实战检验的产品管理技能,由 [Dean Peters](https://github.com/deanpeters). 提供——用于界定问题、发掘机会、搭建验证实验,并快速否决劣质方案——所采用的框架来自 Teresa Torres、Geoffrey Moore、Amazon、MITRE 等。
|
||
|
||
**组件技能**
|
||
|
||
- **[deanpeters/acquisition-channel-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/acquisition-channel-advisor)** - 基于单位经济模型评估渠道,并给出扩量/测试/终止建议
|
||
- **[deanpeters/ai-shaped-readiness-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/ai-shaped-readiness-advisor)** - 从五项能力维度评估自动化与重新设计的机会
|
||
- **[deanpeters/altitude-horizon-framework](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/altitude-horizon-framework)** - 应对 PM 向总监的思维转变,涵盖范围、时间跨度与失败模式
|
||
- **[deanpeters/business-health-diagnostic](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/business-health-diagnostic)** - 诊断 SaaS 健康状况,识别危险信号,并确定恢复行动优先级
|
||
- **[deanpeters/company-research](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/company-research)** - 深入分析竞争对手或公司
|
||
- **[deanpeters/customer-journey-map](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/customer-journey-map)** - 使用 NNGroup 框架映射各触点的客户体验
|
||
- **[deanpeters/eol-message](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/eol-message)** - 得体地传达产品或功能弃用信息
|
||
- **[deanpeters/epic-hypothesis](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/epic-hypothesis)** - 将举措转化为可测试假设,并设定可衡量的成功指标
|
||
- **[deanpeters/finance-metrics-quickref](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/finance-metrics-quickref)** - 32+ 项 SaaS 财务指标参考指南,含公式与基准
|
||
- **[deanpeters/jobs-to-be-done](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/jobs-to-be-done)** - 使用 JTBD(Jobs-to-be-Done)框架理解客户目标
|
||
- **[deanpeters/pestel-analysis](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/pestel-analysis)** - 从政治、经济、社会、技术、环境、法律等维度分析外部因素
|
||
- **[deanpeters/pol-probe](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/pol-probe)** - 定义轻量级验证实验以检验假设
|
||
- **[deanpeters/positioning-statement](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/positioning-statement)** - 使用 Geoffrey Moore 框架定义目标受众、解决的问题与差异化
|
||
- **[deanpeters/press-release](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/press-release)** - 运用 Amazon 的 Working Backwards 方法,用一篇未来新闻稿阐明产品愿景
|
||
- **[deanpeters/problem-statement](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/problem-statement)** - 在提出方案前,用证据界定客户问题
|
||
- **[deanpeters/proto-persona](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/proto-persona)** - 在开展全面调研前创建假设驱动的用户画像(persona)
|
||
- **[deanpeters/recommendation-canvas](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/recommendation-canvas)** - 记录 AI 驱动的产品推荐方案
|
||
- **[deanpeters/saas-economics-efficiency-metrics](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/saas-economics-efficiency-metrics)** - 计算单位经济模型与资本效率,包括 CAC、LTV、回本周期与 Rule of 40
|
||
- **[deanpeters/saas-revenue-growth-metrics](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/saas-revenue-growth-metrics)** - 跟踪收入、留存与增长指标,包括 MRR/ARR、流失率、NRR 与扩展收入
|
||
- **[deanpeters/storyboard](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/storyboard)** - 用 6 格叙事故事板可视化用户旅程
|
||
- **[deanpeters/user-story](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/user-story)** - 使用 Mike Cohn 与 Gherkin 格式撰写含验收标准的用户故事
|
||
- **[deanpeters/user-story-mapping](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/user-story-mapping)** - 使用 Jeff Patton 的故事地图法按用户工作流组织故事
|
||
- **[deanpeters/user-story-splitting](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/user-story-splitting)** - 使用 8 种经证实的拆分模式拆解大型故事
|
||
|
||
**交互式技能(Interactive Skills)**
|
||
|
||
- **[deanpeters/context-engineering-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/context-engineering-advisor)** - 诊断 context stuffing(上下文堆砌)与 context engineering(上下文工程)的差异,并指导记忆与检索设计
|
||
- **[deanpeters/customer-journey-mapping-workshop](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/customer-journey-mapping-workshop)** - 引导用户旅程映射工作坊,识别痛点
|
||
- **[deanpeters/director-readiness-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/director-readiness-advisor)** - 在四种关键情境下辅导 PM→Director(产品经理→总监)转型
|
||
- **[deanpeters/discovery-interview-prep](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/discovery-interview-prep)** - 根据研究目标,用 Mom Test 风格规划客户访谈
|
||
- **[deanpeters/epic-breakdown-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/epic-breakdown-advisor)** - 运用 Richard Lawrence 的 9 种拆分模式将史诗拆分为用户故事
|
||
- **[deanpeters/feature-investment-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/feature-investment-advisor)** - 通过 ROI 和战略价值评分评估功能投资
|
||
- **[deanpeters/finance-based-pricing-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/finance-based-pricing-advisor)** - 通过财务影响分析评估定价变更
|
||
- **[deanpeters/lean-ux-canvas](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/lean-ux-canvas)** - 使用 Jeff Gothelf 的 Lean UX Canvas v2 建立假设驱动式规划
|
||
- **[deanpeters/opportunity-solution-tree](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/opportunity-solution-tree)** - 生成机会与解决方案,并推荐概念验证(proof-of-concept)测试
|
||
- **[deanpeters/pol-probe-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/pol-probe-advisor)** - 推荐原型类型:Feasibility、Task-Focused、Narrative、Synthetic 或 Vibe
|
||
- **[deanpeters/positioning-workshop](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/positioning-workshop)** - 通过自适应探索式问题引导定位定义
|
||
- **[deanpeters/prioritization-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/prioritization-advisor)** - 根据你的情境推荐合适的优先级框架(RICE、ICE、Kano 等)
|
||
- **[deanpeters/problem-framing-canvas](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/problem-framing-canvas)** - 引导完成 MITRE Problem Framing:向内看(Look Inward)、向外看(Outward)、再框定(Reframe)
|
||
- **[deanpeters/tam-sam-som-calculator](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/tam-sam-som-calculator)** - 基于真实数据和引用推算市场规模
|
||
- **[deanpeters/user-story-mapping-workshop](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/user-story-mapping-workshop)** - 逐步创建用户故事地图,包含主干(backbone)与发布切片(release slices)
|
||
- **[deanpeters/vp-cpo-readiness-advisor](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/vp-cpo-readiness-advisor)** - 辅导 Director→VP/CPO(总监→副总裁/首席产品官)转型,含 CEO 面试框架
|
||
- **[deanpeters/workshop-facilitation](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/workshop-facilitation)** - 为任意工作坊添加分步引导与编号建议
|
||
|
||
**工作流技能(Workflow Skills)**
|
||
|
||
- **[deanpeters/discovery-process](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/discovery-process)** - 完整发现周期:框定问题 → 调研 → 综合 → 验证(3-4 周)
|
||
- **[deanpeters/executive-onboarding-playbook](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/executive-onboarding-playbook)** - VP/CPO 入职转型的 30-60-90 天诊断手册
|
||
- **[deanpeters/prd-development](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/prd-development)** - 结构化 PRD 流程:问题 → 人物角色(personas)→ 解决方案 → 指标 → 用户故事(2-4 天)
|
||
- **[deanpeters/product-strategy-session](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/product-strategy-session)** - 完整战略会议:定位 → 框定 → 探索 → 路线图(2-4 周)
|
||
- **[deanpeters/roadmap-planning](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/roadmap-planning)** - 战略路线图流程:输入 → 史诗 → 排优先级 → 排序 → 沟通(1-2 周)
|
||
- **[deanpeters/skill-authoring-workflow](https://github.com/deanpeters/Product-Manager-Skills/tree/main/skills/skill-authoring-workflow)** - 技能编写的元流程:选择路径 → 验证 → 更新文档 → 打包
|
||
|
||
</details>
|
||
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Pawel Huryn 的产品管理技能</h3></summary>
|
||
|
||
由 [Paweł Huryn](https://github.com/phuryn), creator of The Product Compass newsletter. 创作的 65 项产品管理技能,The Product Compass 通讯创办人。覆盖完整 PM 生命周期——从发现与战略到执行、分析和 go-to-market(上市)——并融入 Teresa Torres、Geoffrey Moore 等人的框架。
|
||
|
||
**数据分析(Data Analytics)**
|
||
|
||
- **[phuryn/ab-test-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-data-analytics/skills/ab-test-analysis)** - 分析 A/B 测试结果,含统计显著性与建议
|
||
- **[phuryn/cohort-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-data-analytics/skills/cohort-analysis)** - 队列留存曲线、功能采纳与细分洞察
|
||
- **[phuryn/sql-queries](https://github.com/phuryn/pm-skills/tree/main/pm-data-analytics/skills/sql-queries)** - 从自然语言生成主要方言的 SQL 查询
|
||
|
||
**执行(Execution)**
|
||
|
||
- **[phuryn/brainstorm-okrs](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/brainstorm-okrs)** - 头脑风暴与公司目标对齐的团队 OKR
|
||
- **[phuryn/create-prd](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/create-prd)** - 用 8 节模板创建 PRD,覆盖从问题到发布
|
||
- **[phuryn/dummy-dataset](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/dummy-dataset)** - 生成 CSV、JSON 或 SQL 格式的逼真虚拟数据集
|
||
- **[phuryn/job-stories](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/job-stories)** - 用 JTBD 格式创建带验收标准的工作故事
|
||
- **[phuryn/outcome-roadmap](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/outcome-roadmap)** - 将产出型路线图转化为以成果为中心的战略计划
|
||
- **[phuryn/pre-mortem](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/pre-mortem)** - 对 PRD 和发布计划进行事前验尸(pre-mortem)风险分析
|
||
- **[phuryn/prioritization-frameworks](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/prioritization-frameworks)** - 含模板的 9 种优先级框架参考指南
|
||
- **[phuryn/release-notes](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/release-notes)** - 从工单或变更日志生成面向用户的发布说明
|
||
- **[phuryn/retro](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/retro)** - 引导结构化冲刺回顾并产出行动项
|
||
- **[phuryn/sprint-plan](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/sprint-plan)** - 规划冲刺,含产能、故事选择与风险映射
|
||
- **[phuryn/stakeholder-map](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/stakeholder-map)** - 构建利益相关者地图,含权力/利益矩阵与沟通计划
|
||
- **[phuryn/summarize-meeting](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/summarize-meeting)** - 将会议记录总结为结构化笔记与行动项
|
||
- **[phuryn/test-scenarios](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/test-scenarios)** - 从用户故事创建全面测试场景
|
||
- **[phuryn/user-stories](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/user-stories)** - 创建符合 INVEST 原则、采用 3 C's 结构的用户故事
|
||
- **[phuryn/wwas](https://github.com/phuryn/pm-skills/tree/main/pm-execution/skills/wwas)** - 以 Why-What-Acceptance 格式创建待办项
|
||
|
||
**市场进入(Go-to-Market)**
|
||
|
||
- **[phuryn/beachhead-segment](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/beachhead-segment)** - 识别产品发布的首个滩头阵地市场细分
|
||
- **[phuryn/competitive-battlecard](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/competitive-battlecard)** - 针对特定竞争对手创建可直接用于销售的竞争情报卡(battlecard)
|
||
- **[phuryn/growth-loops](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/growth-loops)** - 识别 5 种飞轮类型中的增长闭环以获取牵引力
|
||
- **[phuryn/gtm-motions](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/gtm-motions)** - 识别包括 PLG 和 ABM 在内的 7 种最佳 GTM 模式
|
||
- **[phuryn/gtm-strategy](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/gtm-strategy)** - 制定包含渠道、信息和发布时间线的 GTM 策略
|
||
- **[phuryn/ideal-customer-profile](https://github.com/phuryn/pm-skills/tree/main/pm-go-to-market/skills/ideal-customer-profile)** - 识别包含人口统计、行为和 JTBD 的 ICP
|
||
|
||
**市场研究(Market Research)**
|
||
|
||
- **[phuryn/competitor-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/competitor-analysis)** - 分析竞争对手的优势、劣势和差异化
|
||
- **[phuryn/customer-journey-map](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/customer-journey-map)** - 绘制包含触点、情绪和机会的客户旅程地图
|
||
- **[phuryn/market-segments](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/market-segments)** - 识别 3-5 个包含 JTBD 和产品契合度的客户细分
|
||
- **[phuryn/market-sizing](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/market-sizing)** - 使用自上而下和自下而上方法估算 TAM、SAM、SOM
|
||
- **[phuryn/sentiment-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/sentiment-analysis)** - 分析用户反馈的情感得分和 JTBD 洞察
|
||
- **[phuryn/user-personas](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/user-personas)** - 创建 3 个包含 JTBD、痛点和收益的用户画像
|
||
- **[phuryn/user-segmentation](https://github.com/phuryn/pm-skills/tree/main/pm-market-research/skills/user-segmentation)** - 根据反馈数据按行为、JTBD 和需求对用户进行分群
|
||
|
||
**营销与增长(Marketing & Growth)**
|
||
|
||
- **[phuryn/marketing-ideas](https://github.com/phuryn/pm-skills/tree/main/pm-marketing-growth/skills/marketing-ideas)** - 生成 5 个创意十足、高性价比的营销方案及理由
|
||
- **[phuryn/north-star-metric](https://github.com/phuryn/pm-skills/tree/main/pm-marketing-growth/skills/north-star-metric)** - 定义北极星指标(North Star Metric)及输入指标群
|
||
- **[phuryn/positioning-ideas](https://github.com/phuryn/pm-skills/tree/main/pm-marketing-growth/skills/positioning-ideas)** - 头脑风暴与竞争对手差异化的定位创意
|
||
- **[phuryn/product-name](https://github.com/phuryn/pm-skills/tree/main/pm-marketing-growth/skills/product-name)** - 头脑风暴 5 个契合品牌价值观、令人难忘的产品名称
|
||
- **[phuryn/value-prop-statements](https://github.com/phuryn/pm-skills/tree/main/pm-marketing-growth/skills/value-prop-statements)** - 为营销、销售和用户上手生成价值主张陈述
|
||
|
||
**产品探索(Product Discovery)**
|
||
|
||
- **[phuryn/analyze-feature-requests](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/analyze-feature-requests)** - 按主题、影响、投入和风险对功能需求进行优先级排序
|
||
- **[phuryn/brainstorm-experiments-existing](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/brainstorm-experiments-existing)** - 为现有产品设计实验以验证假设
|
||
- **[phuryn/brainstorm-experiments-new](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/brainstorm-experiments-new)** - 设计精益原型测试(pretotype)以验证新产品
|
||
- **[phuryn/brainstorm-ideas-existing](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/brainstorm-ideas-existing)** - 从 PM、设计师、工程师视角头脑风暴产品创意
|
||
- **[phuryn/brainstorm-ideas-new](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/brainstorm-ideas-new)** - 在早期探索阶段为新产品头脑风暴功能创意
|
||
- **[phuryn/identify-assumptions-existing](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/identify-assumptions-existing)** - 识别价值、可用性、可行性、可落地性方面的风险假设
|
||
- **[phuryn/identify-assumptions-new](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/identify-assumptions-new)** - 识别新产品在 8 个风险类别中的风险假设
|
||
- **[phuryn/interview-script](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/interview-script)** - 创建包含 JTBD 探询的结构化客户访谈脚本
|
||
- **[phuryn/metrics-dashboard](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/metrics-dashboard)** - 定义包含数据源和告警阈值的产品指标看板
|
||
- **[phuryn/opportunity-solution-tree](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/opportunity-solution-tree)** - 基于 Teresa Torres 的方法构建机会解决方案树(Opportunity Solution Tree)
|
||
- **[phuryn/prioritize-assumptions](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/prioritize-assumptions)** - 使用影响 × 风险矩阵和实验对假设进行优先级排序
|
||
- **[phuryn/prioritize-features](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/prioritize-features)** - 按影响、投入、风险和战略契合度对待办项进行优先级排序
|
||
- **[phuryn/summarize-interview](https://github.com/phuryn/pm-skills/tree/main/pm-product-discovery/skills/summarize-interview)** - 总结访谈记录,提炼 JTBD 和行动项
|
||
|
||
**产品战略(Product Strategy)**
|
||
|
||
- **[phuryn/ansoff-matrix](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/ansoff-matrix)** - 安索夫矩阵(Ansoff Matrix)四象限增长战略分析
|
||
- **[phuryn/business-model](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/business-model)** - 生成包含全部 9 个构建模块的商业模式画布(Business Model Canvas)
|
||
- **[phuryn/lean-canvas](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/lean-canvas)** - 生成包含问题、解决方案、UVP 和指标的精益画布(Lean Canvas)
|
||
- **[phuryn/monetization-strategy](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/monetization-strategy)** - 头脑风暴 3-5 种变现策略及验证实验
|
||
- **[phuryn/pestle-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/pestle-analysis)** - 针对政治、经济、社会、技术、法律、环境维度的 PESTLE 分析
|
||
- **[phuryn/porters-five-forces](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/porters-five-forces)** - 波特五力(Porter's Five Forces)竞争分析及战略洞察
|
||
- **[phuryn/pricing-strategy](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/pricing-strategy)** - 设计包含竞争分析和 WTP(支付意愿)估算的定价策略
|
||
- **[phuryn/product-strategy](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/product-strategy)** - 使用 9 部分产品战略画布创建产品战略
|
||
- **[phuryn/product-vision](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/product-vision)** - 头脑风暴鼓舞人心、可实现的产品愿景陈述
|
||
- **[phuryn/startup-canvas](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/startup-canvas)** - 生成结合产品战略与商业模式的创业画布(Startup Canvas)
|
||
- **[phuryn/swot-analysis](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/swot-analysis)** - SWOT 分析及各象限可执行建议
|
||
- **[phuryn/value-proposition](https://github.com/phuryn/pm-skills/tree/main/pm-product-strategy/skills/value-proposition)** - 使用 6 部分 JTBD 模板设计价值主张
|
||
|
||
**工具包**
|
||
|
||
- **[phuryn/draft-nda](https://github.com/phuryn/pm-skills/tree/main/pm-toolkit/skills/draft-nda)** - 起草保密协议(NDA),涵盖信息类型、司法管辖区和条款
|
||
- **[phuryn/grammar-check](https://github.com/phuryn/pm-skills/tree/main/pm-toolkit/skills/grammar-check)** - 识别语法与行文流畅性问题,并提供针对性修改建议
|
||
- **[phuryn/privacy-policy](https://github.com/phuryn/pm-skills/tree/main/pm-toolkit/skills/privacy-policy)** - 起草隐私政策,并考虑 GDPR 合规要求
|
||
- **[phuryn/review-resume](https://github.com/phuryn/pm-skills/tree/main/pm-toolkit/skills/review-resume)** - 对照 10 项最佳实践(含 XYZ+S 公式)审阅产品经理简历
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">MiniMax 团队的技能</h3></summary>
|
||
|
||
MiniMax AI 团队提供的 11 项开发与文档生成技能。覆盖前端、全栈、移动端、着色器(shader)开发、文档创建(PDF、DOCX、XLSX、PPTX),以及集成 MiniMax API 的 MiniMax AI CLI。
|
||
|
||
- **[MiniMax-AI/cli](https://officialskills.sh/MiniMax-AI/skills/cli)** - 通过 MiniMax AI 生成文本、图像、视频、语音和音乐
|
||
- **[MiniMax-AI/frontend-dev](https://officialskills.sh/MiniMax-AI/skills/frontend-dev)** - 全栈前端开发,含电影级动画、通过 MiniMax API 生成 AI 媒体内容,以及生成式艺术
|
||
- **[MiniMax-AI/fullstack-dev](https://officialskills.sh/MiniMax-AI/skills/fullstack-dev)** - 后端架构,含 REST API 设计、认证流程、实时功能与数据库集成
|
||
- **[MiniMax-AI/android-native-dev](https://officialskills.sh/MiniMax-AI/skills/android-native-dev)** - Android 原生开发,使用 Kotlin/Jetpack Compose、Material Design 3 与无障碍支持
|
||
- **[MiniMax-AI/ios-application-dev](https://officialskills.sh/MiniMax-AI/skills/ios-application-dev)** - iOS 开发,使用 UIKit、SnapKit 与 SwiftUI,覆盖导航、深色模式与 HIG 合规
|
||
- **[MiniMax-AI/shader-dev](https://officialskills.sh/MiniMax-AI/skills/shader-dev)** - GLSL 着色器技术,用于光线步进、流体模拟、粒子系统与程序化生成
|
||
- **[MiniMax-AI/gif-sticker-maker](https://officialskills.sh/MiniMax-AI/skills/gif-sticker-maker)** - 通过 MiniMax API 将照片转换为 Funko Pop / Pop Mart 风格的动画 GIF 贴纸
|
||
- **[MiniMax-AI/minimax-pdf](https://officialskills.sh/MiniMax-AI/skills/minimax-pdf)** - 基于令牌的设计系统生成、填写与重排 PDF,提供 15 种封面样式
|
||
- **[MiniMax-AI/pptx-generator](https://officialskills.sh/MiniMax-AI/skills/pptx-generator)** - 使用 PptxGenJS 从零创建与编辑 PowerPoint 演示文稿
|
||
- **[MiniMax-AI/minimax-xlsx](https://officialskills.sh/MiniMax-AI/skills/minimax-xlsx)** - 创建、读取、分析与校验 Excel/电子表格文件,零格式损失
|
||
- **[MiniMax-AI/minimax-docx](https://officialskills.sh/MiniMax-AI/skills/minimax-docx)** - 使用 OpenXML SDK 进行专业 DOCX 文档创建与编辑
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">DuckDB 的技能</h3></summary>
|
||
|
||
官方 DuckDB 技能,用于数据查询、文件读取,以及直接在 Claude Code 中搜索文档。
|
||
|
||
- **[duckdb/attach-db](https://officialskills.sh/duckdb/skills/attach-db)** - 附加 DuckDB 数据库文件以进行交互式查询,并自动探索 schema
|
||
- **[duckdb/query](https://officialskills.sh/duckdb/skills/query)** - 对已附加数据库或临时文件运行 SQL 查询,使用 Friendly SQL 方言
|
||
- **[duckdb/read-file](https://officialskills.sh/duckdb/skills/read-file)** - 读取任意数据文件(CSV、JSON、Parquet、Avro、Excel、空间数据),支持本地或远程存储
|
||
- **[duckdb/duckdb-docs](https://officialskills.sh/duckdb/skills/duckdb-docs)** - 通过 HTTPS 全文搜索 DuckDB 与 DuckLake 文档
|
||
- **[duckdb/read-memories](https://officialskills.sh/duckdb/skills/read-memories)** - 搜索过往 Claude Code 会话日志,从先前对话中恢复上下文
|
||
- **[duckdb/install-duckdb](https://officialskills.sh/duckdb/skills/install-duckdb)** - 安装或更新 DuckDB CLI 与扩展,并进行版本管理
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">GSAP (GreenSock) 的技能</h3></summary>
|
||
|
||
官方 GSAP 动画技能,覆盖完整 GreenSock 生态——核心 API、时间轴、ScrollTrigger、插件、工具函数、React 集成、性能优化与框架支持。
|
||
|
||
- **[greensock/gsap-core](https://officialskills.sh/greensock/skills/gsap-core)** - 核心 API:gsap.to()、from()、fromTo()、缓动、easing、时长、stagger 与默认值
|
||
- **[greensock/gsap-timeline](https://officialskills.sh/greensock/skills/gsap-timeline)** - 时间轴:序列编排、position 参数、标签、嵌套与播放控制
|
||
- **[greensock/gsap-scrolltrigger](https://officialskills.sh/greensock/skills/gsap-scrolltrigger)** - ScrollTrigger:滚动联动动画、固定(pinning)、scrub 与刷新处理
|
||
- **[greensock/gsap-plugins](https://officialskills.sh/greensock/skills/gsap-plugins)** - 插件:ScrollToPlugin、Flip、Draggable、SplitText、SVG 与物理效果
|
||
- **[greensock/gsap-utils](https://officialskills.sh/greensock/skills/gsap-utils)** - 工具函数:clamp、mapRange、interpolate、snap、selector、wrap
|
||
- **[greensock/gsap-react](https://officialskills.sh/greensock/skills/gsap-react)** - React 集成:useGSAP hook、refs、gsap.context()、清理与 SSR
|
||
- **[greensock/gsap-performance](https://officialskills.sh/greensock/skills/gsap-performance)** - 性能建议:transforms、will-change、批处理与 ScrollTrigger 优化
|
||
- **[greensock/gsap-frameworks](https://officialskills.sh/greensock/skills/gsap-frameworks)** - Vue、Svelte 等框架的生命周期、作用域与清理模式
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Garry Tan (gstack) 的技能</h3></summary>
|
||
|
||
[Garry Tan](https://github.com/garrytan)(Y Combinator CEO)提供的 28 项技能,将 Claude Code 转变为虚拟工程团队——从创意到生产部署的结构化工作流。60 天内交付 60 万+ 行生产代码。
|
||
|
||
- **[garrytan/office-hours](https://officialskills.sh/garrytan/skills/office-hours)** - YC Office Hours:六个强制性问题,在写代码前重新框定你的产品
|
||
- **[garrytan/plan-ceo-review](https://officialskills.sh/garrytan/skills/plan-ceo-review)** - CEO/创始人计划审阅,四种模式:Expansion、Selective Expansion、Hold Scope、Reduction
|
||
- **[garrytan/plan-eng-review](https://officialskills.sh/garrytan/skills/plan-eng-review)** - 工程经理审阅:锁定架构、数据流、图表、边界情况与测试
|
||
- **[garrytan/plan-design-review](https://officialskills.sh/garrytan/skills/plan-design-review)** - 资深设计师审阅:各设计维度 0-10 分评分,说明 10 分标准,AI Slop 检测
|
||
- **[garrytan/design-consultation](https://officialskills.sh/garrytan/skills/design-consultation)** - 从零构建完整设计系统,含创意风险与逼真产品 mockup
|
||
- **[garrytan/design-review](https://officialskills.sh/garrytan/skills/design-review)** - Designer Who Codes:视觉审计,随后以原子提交与前后对比截图进行修复
|
||
- **[garrytan/review](https://officialskills.sh/garrytan/skills/review)** - Staff Engineer 代码审阅:发现能通过 CI 但在生产环境崩溃的 bug
|
||
- **[garrytan/investigate](https://officialskills.sh/garrytan/skills/investigate)** - 系统化根因调试:无调查不修复,追踪数据流,验证假设
|
||
- **[garrytan/qa](https://officialskills.sh/garrytan/skills/qa)** - QA Lead:测试应用、发现 bug、以原子提交修复,并自动生成回归测试
|
||
- **[garrytan/qa-only](https://officialskills.sh/garrytan/skills/qa-only)** - QA Reporter:与 /qa 相同方法论,但仅报告、不修改代码
|
||
- **[garrytan/cso](https://officialskills.sh/garrytan/skills/cso)** - Chief Security Officer:OWASP Top 10 + STRIDE 威胁模型,零误报排除
|
||
- **[garrytan/ship](https://officialskills.sh/garrytan/skills/ship)** - Release Engineer:同步 main、运行测试、审计覆盖率、推送并打开 PR
|
||
- **[garrytan/land-and-deploy](https://officialskills.sh/garrytan/skills/land-and-deploy)** - 合并 PR、等待 CI 与部署、验证生产环境健康状态
|
||
- **[garrytan/canary](https://officialskills.sh/garrytan/skills/canary)** - SRE 部署后监控:关注控制台错误、性能回退与页面故障
|
||
- **[garrytan/benchmark](https://officialskills.sh/garrytan/skills/benchmark)** - Performance Engineer:基线页面加载时间、Core Web Vitals 与资源体积
|
||
- **[garrytan/document-release](https://officialskills.sh/garrytan/skills/document-release)** - Technical Writer:更新所有项目文档以匹配刚发布的内容
|
||
- **[garrytan/retro](https://officialskills.sh/garrytan/skills/retro)** - 工程经理周度复盘,含按人分解与连续交付记录
|
||
- **[garrytan/browse](https://officialskills.sh/garrytan/skills/browse)** - 真实 Chromium 浏览器用于 QA:真实点击、真实截图,每条命令约 100ms
|
||
- **[garrytan/setup-browser-cookies](https://officialskills.sh/garrytan/skills/setup-browser-cookies)** - 将真实浏览器的 cookie 导入无头会话
|
||
- **[garrytan/autoplan](https://officialskills.sh/garrytan/skills/autoplan)** - 一条命令生成完整审阅计划:自动运行 CEO → 设计 → 工程审阅
|
||
- **[garrytan/codex](https://officialskills.sh/garrytan/skills/codex)** - 通过 OpenAI Codex CLI 获取第二意见:审阅、对抗性挑战与开放咨询
|
||
- **[garrytan/careful](https://officialskills.sh/garrytan/skills/careful)** - Safety Guardrails:在破坏性命令(rm -rf、DROP TABLE、force-push)前发出警告
|
||
- **[garrytan/freeze](https://officialskills.sh/garrytan/skills/freeze)** - Edit Lock:调试时将文件编辑限制在单一目录
|
||
- **[garrytan/guard](https://officialskills.sh/garrytan/skills/guard)** - Full Safety:一条命令同时启用 /careful + /freeze,最大化安全
|
||
- **[garrytan/unfreeze](https://officialskills.sh/garrytan/skills/unfreeze)** - Unlock:移除 /freeze 边界限制
|
||
- **[garrytan/setup-deploy](https://officialskills.sh/garrytan/skills/setup-deploy)** - Deploy Configurator:为 /land-and-deploy 进行一次性配置
|
||
- **[garrytan/gstack-upgrade](https://officialskills.sh/garrytan/skills/gstack-upgrade)** - Self-Updater:将 gstack 升级到最新版本
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Notion 出品的 Skills</h3></summary>
|
||
|
||
来自 Notion 官方仓库的技能 — 面向工作区的技能,可用于知识沉淀、会议准备、调研,并将规格说明转化为任务。
|
||
|
||
**来自 [notion-cookbook](https://github.com/makenotion/notion-cookbook/tree/main/skills/claude):**
|
||
|
||
- **[makenotion/knowledge-capture](https://officialskills.sh/makenotion/skills/knowledge-capture)** - 将对话转化为结构化的 Notion 文档页面,并做好组织与链接
|
||
- **[makenotion/meeting-intelligence](https://officialskills.sh/makenotion/skills/meeting-intelligence)** - 通过收集 Notion 上下文,准备会议材料、预读资料和议程
|
||
- **[makenotion/research-documentation](https://officialskills.sh/makenotion/skills/research-documentation)** - 搜索 Notion 工作区,综合调研发现,并创建全面的研究报告
|
||
- **[makenotion/spec-to-implementation](https://officialskills.sh/makenotion/skills/spec-to-implementation)** - 将产品/技术规格转化为具体的 Notion 任务,附带验收标准与进度跟踪
|
||
|
||
**来自 [claude-code-notion-plugin](https://github.com/makenotion/claude-code-notion-plugin/tree/main/skills/notion):**
|
||
|
||
- **[makenotion/knowledge-capture](https://officialskills.sh/makenotion/skills/knowledge-capture)** - 将对话转化为结构化的 Notion 文档页面,并做好组织与链接
|
||
- **[makenotion/meeting-intelligence](https://officialskills.sh/makenotion/skills/meeting-intelligence)** - 通过收集 Notion 上下文,准备会议材料、预读资料和议程
|
||
- **[makenotion/research-documentation](https://officialskills.sh/makenotion/skills/research-documentation)** - 搜索 Notion 工作区,综合调研发现,并创建全面的研究报告
|
||
- **[makenotion/spec-to-implementation](https://officialskills.sh/makenotion/skills/spec-to-implementation)** - 将产品/技术规格转化为具体的 Notion 任务,附带验收标准与进度跟踪
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Resend 出品的 Skills</h3></summary>
|
||
|
||
官方 Resend 技能,用于收发邮件、构建邮件模板,并为你的智能体注入邮件专业能力。
|
||
|
||
- **[resend/resend](https://github.com/resend/resend-skills/tree/main/skills/resend)** - 通过 Resend API 发送和管理邮件
|
||
- **[resend/react-email](https://github.com/resend/resend-skills/tree/main/skills/react-email)** - 使用 React Email 组件构建邮件
|
||
- **[resend/email-best-practices](https://github.com/resend/resend-skills/tree/main/skills/email-best-practices)** - 邮件送达率与设计最佳实践
|
||
- **[resend/agent-email-inbox](https://github.com/resend/resend-skills/tree/main/skills/agent-email-inbox)** - AI 智能体邮件收件箱管理
|
||
- **[resend/resend-cli](https://github.com/resend/resend-skills/tree/main/skills/resend-cli)** - Resend CLI 命令与工作流
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Google Chrome 团队 — Addy Osmani(Web Quality)出品的 Skills</h3></summary>
|
||
|
||
来自 Addy Osmani(Google Chrome 团队)的 Lighthouse 风格 Web 质量技能,涵盖性能、Core Web Vitals、无障碍、SEO 与现代最佳实践 — 与 Google Lighthouse 审计的支柱一致。
|
||
|
||
- **[addyosmani/web-quality-audit](https://officialskills.sh/addyosmani/skills/web-quality-audit)** - 全面质量审查,涵盖性能、无障碍、SEO 与最佳实践类别
|
||
- **[addyosmani/performance](https://officialskills.sh/addyosmani/skills/performance)** - 加载速度、运行时效率与资源优化
|
||
- **[addyosmani/core-web-vitals](https://officialskills.sh/addyosmani/skills/core-web-vitals)** - LCP、INP 和 CLS 专项优化
|
||
- **[addyosmani/accessibility](https://officialskills.sh/addyosmani/skills/accessibility)** - WCAG 合规、屏幕阅读器支持与键盘导航
|
||
- **[addyosmani/seo](https://officialskills.sh/addyosmani/skills/seo)** - 搜索引擎优化、可爬取性与结构化数据
|
||
- **[addyosmani/best-practices](https://officialskills.sh/addyosmani/skills/best-practices)** - 安全性、现代 Web API 与代码质量模式
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">MongoDB 出品的 Skills</h3></summary>
|
||
|
||
面向智能体工作流的官方 MongoDB Agent Skills — 连接管理、schema 设计、查询优化、自然语言查询,以及 Atlas Stream Processing。
|
||
|
||
- **[mongodb/mongodb-mcp-setup](https://officialskills.sh/mongodb/skills/mongodb-mcp-setup)** - 配置 MongoDB MCP 服务器,包括认证与连接设置
|
||
- **[mongodb/mongodb-connection](https://officialskills.sh/mongodb/skills/mongodb-connection)** - 优化 MongoDB 客户端连接池、超时与 serverless 模式
|
||
- **[mongodb/mongodb-schema-design](https://officialskills.sh/mongodb/skills/mongodb-schema-design)** - 设计高效的文档 schema,含验证与索引模式
|
||
- **[mongodb/atlas-stream-processing](https://officialskills.sh/mongodb/skills/atlas-stream-processing)** - 构建、运维和调试 Atlas Stream Processing 流水线,集成 Kafka、S3 与 Lambda
|
||
- **[mongodb/mongodb-natural-language-querying](https://officialskills.sh/mongodb/skills/mongodb-natural-language-querying)** - 将自然语言转化为 MongoDB 查询与聚合管道
|
||
- **[mongodb/mongodb-query-optimizer](https://officialskills.sh/mongodb/skills/mongodb-query-optimizer)** - 使用 Atlas Performance Advisor 分析与优化查询性能
|
||
- **[mongodb/mongodb-search-and-ai](https://officialskills.sh/mongodb/skills/mongodb-search-and-ai)** - 实现 Atlas Search 与 AI 推荐,含向量搜索
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Redis 出品的 Skills</h3></summary>
|
||
|
||
- **[redis/redis-development](https://github.com/redis/agent-skills/tree/main/skills/redis-development)** - Redis 开发最佳实践 — 数据结构、查询引擎、向量搜索、缓存与性能优化。
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">NVIDIA 出品的 Skills</h3></summary>
|
||
|
||
NVIDIA 在 AI/ML 基础设施项目中发布的官方技能 — Megatron、NeMo、TensorRT-LLM、CUDA-Q、cuOpt、DALI、DeepStream 等。17 款产品共 155 个技能。
|
||
|
||
### CUDA-Q
|
||
|
||
- **[NVIDIA/CUDA-Q/cudaq-guide](https://github.com/NVIDIA/skills/tree/main/skills/CUDA-Q/cudaq-guide)** - CUDA-Q 入门指南,涵盖安装、测试程序、GPU 模拟、QPU 硬件与量子应用。
|
||
|
||
### DALI
|
||
|
||
- **[NVIDIA/DALI/dali-dynamic-mode](https://github.com/NVIDIA/skills/tree/main/skills/DALI/dali-dynamic-mode)** - 在编写 DALI 数据加载或预处理代码且使用 `nvidia.dali.experimental.dynamic` (ndd),或将 DALI pipeline 模式代码转换为 dynamic 模式时使用,或当用户...
|
||
|
||
### Megatron-Bridge
|
||
|
||
- **[NVIDIA/Megatron-Bridge/adding-model-support](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/adding-model-support)** - 在 Megatron-Bridge 中添加新 LLM 或 VLM 模型支持的指南。
|
||
- **[NVIDIA/Megatron-Bridge/build-and-dependency](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/build-and-dependency)** - Megatron Bridge 开发环境搭建 — 基于容器的开发、uv 包管理、lockfile 重新生成、添加依赖、Slurm 容器使用,以及常见构建...
|
||
- **[NVIDIA/Megatron-Bridge/bump-dependency](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/bump-dependency)** - 升级固定版本依赖(TransformerEngine、Megatron-LM、NRX 等),重新生成 lockfile,提交 PR,并通过挂载 watchdog 到「CICD NeMo」工作流... 直至绿灯
|
||
- **[NVIDIA/Megatron-Bridge/cicd](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/cicd)** - Megatron Bridge 的 CI/CD 参考 — 流水线结构、提交与 PR 工作流、CI 失败排查,以及常见失败模式。
|
||
- **[NVIDIA/Megatron-Bridge/linting-and-formatting](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/linting-and-formatting)** - Megatron Bridge 代码风格与质量规则 — ruff 配置、命名约定、类型提示、mypy 规则、文档字符串、版权头、日志,以及代码审查检查...
|
||
- **[NVIDIA/Megatron-Bridge/mlm-bridge-training](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/mlm-bridge-training)** - 使用 mock 或真实数据运行 Megatron-LM (MLM) 与 Megatron Bridge 训练。
|
||
- **[NVIDIA/Megatron-Bridge/multi-node-slurm](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/multi-node-slurm)** - 将单节点脚本转换为多节点 Slurm sbatch 作业,并调试常见多节点故障。
|
||
- **[NVIDIA/Megatron-Bridge/nemo-rl-e2e-testing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/nemo-rl-e2e-testing)** - 针对 Megatron-Bridge 模型/提供商变更的外部 NeMo-RL 端到端验证工作流,包括下游兼容性检查、外部 RL 生命周期行为、Megatron poli...
|
||
- **[NVIDIA/Megatron-Bridge/parity-testing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/parity-testing)** - 用于验证 HF<->MCore 权重转换数值一致性的结构化框架。
|
||
- **[NVIDIA/Megatron-Bridge/perf-activation-recompute](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-activation-recompute)** - 在 Megatron Bridge 中验证并使用选择性及全量激活重计算,以额外计算换取降低 GPU 显存占用。
|
||
- **[NVIDIA/Megatron-Bridge/perf-cpu-offloading](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-cpu-offloading)** - 在 Megatron Bridge 中验证并使用 CPU offloading,包括层级激活 offloading 以及使用 HybridDeviceOptimizer 的部分优化器状态 offloading。
|
||
- **[NVIDIA/Megatron-Bridge/perf-cuda-graphs](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-cuda-graphs)** - 在 Megatron Bridge 中验证并使用 CUDA graph capture,包括本地全迭代 graph,以及针对 attention、MLP、MoE 模块的 Transformer Engine 作用域 graph。
|
||
- **[NVIDIA/Megatron-Bridge/perf-expert-parallel-overlap](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-expert-parallel-overlap)** - 在 Megatron-Bridge 中验证并使用 MoE expert-parallel 通信重叠,包括 overlap_moe_expert_parallel_comm、delay_wgrad_compute,以及 flex dispatcher 后端等...
|
||
- **[NVIDIA/Megatron-Bridge/perf-hierarchical-context-parallel](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-hierarchical-context-parallel)** - 在 Megatron-Bridge 中启用分层上下文并行的运维指南,包括配置项、代码锚点、陷阱与验证。
|
||
- **[NVIDIA/Megatron-Bridge/perf-megatron-fsdp](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-megatron-fsdp)** - 在 Megatron-Bridge 中启用 Megatron FSDP 的运维指南,包括配置项、代码锚点、陷阱与验证。
|
||
- **[NVIDIA/Megatron-Bridge/perf-memory-tuning](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-memory-tuning)** - 降低 Megatron Bridge 峰值 GPU 显存的技术 — expandable segments、并行度调整、激活重计算、CPU offloading 约束,以及常见 OOM 修复。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-comm-overlap](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-comm-overlap)** - Megatron Bridge 中的 MoE expert-parallel 通信重叠。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-dispatcher-selection](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-dispatcher-selection)** - 根据硬件、EP degree 与优化阶段选择正确的 MoE token dispatcher(`alltoall`、DeepEP 或 HybridEP)。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-hardware-configs](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-hardware-configs)** - 按硬件平台与模型族整理的代表性 MoE 训练 playbook。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-long-context](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-long-context)** - Megatron Bridge 的长上下文 MoE 训练指南。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-optimization-workflow](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-optimization-workflow)** - 基于 Megatron-Core MoE 论文的 Megatron Bridge MoE 训练优化系统化工作流。
|
||
- **[NVIDIA/Megatron-Bridge/perf-moe-vlm-training](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-moe-vlm-training)** - 在 Megatron Bridge 中训练 MoE VLM 的实用指南。
|
||
- **[NVIDIA/Megatron-Bridge/perf-parallelism-strategies](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-parallelism-strategies)** - 在 Megatron Bridge 中选择与组合并行策略的运维指南,包括规模规则、硬件拓扑映射,以及组合并行配置。
|
||
- **[NVIDIA/Megatron-Bridge/perf-sequence-packing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-sequence-packing)** - 在 Megatron-Bridge 中验证并使用 packed sequences 与长上下文训练,区分 LLM 的离线 packed SFT 与 VLM 的 batch 内 packing,并应用正确的 CP...
|
||
- **[NVIDIA/Megatron-Bridge/perf-tp-dp-comm-overlap](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/perf-tp-dp-comm-overlap)** - 在 Megatron-Bridge 中启用 TP、DP 与 PP 通信重叠的运维指南,包括配置项、代码锚点、陷阱与验证。
|
||
- **[NVIDIA/Megatron-Bridge/recipe-recommender](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/recipe-recommender)** - 根据用户的模型、GPU 数量与训练目标推荐并定制 Megatron Bridge recipe。
|
||
- **[NVIDIA/Megatron-Bridge/resiliency](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/resiliency)** - Megatron Bridge 的弹性特性,包括容错、慢节点检测、进程内重启、抢占,以及重跑状态机。
|
||
- **[NVIDIA/Megatron-Bridge/testing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/testing)** - Megatron Bridge 测试参考 — 单元与功能测试布局、层级语义(L0/L1/L2/flaky)、脚本约定、本地运行测试、添加/移动/禁用测试...
|
||
- **[NVIDIA/Megatron-Bridge/verl-e2e-testing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Bridge/verl-e2e-testing)** - 针对 Megatron-Bridge 模型/提供商变更的外部 verl 端到端验证工作流。
|
||
|
||
### Megatron-Core
|
||
|
||
- **[NVIDIA/Megatron-Core/build-and-dependency](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/build-and-dependency)** - 基于容器的开发环境搭建与 Megatron-LM 依赖管理。
|
||
- **[NVIDIA/Megatron-Core/bump-base-image](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/bump-base-image)** - 升级 Megatron-LM CI 使用的 NVIDIA PyTorch 基础镜像 (`nvcr.io/nvidia/pytorch:<YY.MM>-py3`)。
|
||
- **[NVIDIA/Megatron-Core/cicd](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/cicd)** - Megatron-LM 的 CI/CD 参考指南。
|
||
- **[NVIDIA/Megatron-Core/create-issue](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/create-issue)** - 调查失败的 GitHub Actions 运行或作业,并为该失败创建 GitHub issue。
|
||
- **[NVIDIA/Megatron-Core/linting-and-formatting](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/linting-and-formatting)** - Megatron-LM 的 Lint 与格式化。
|
||
- **[NVIDIA/Megatron-Core/nightly-sync](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/nightly-sync)** - 每晚 main 到 dev 同步工作流的领域知识。
|
||
- **[NVIDIA/Megatron-Core/onboard-gb200-1node-tests](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/onboard-gb200-1node-tests)** - 从现有的 MR 范围 2 节点测试接入 GB200 的 1 节点 GitHub MR 功能测试。
|
||
- **[NVIDIA/Megatron-Core/respond-to-issue](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/respond-to-issue)** - 研究并起草对外部贡献者 GitHub issue 或问题的回复。
|
||
- **[NVIDIA/Megatron-Core/run-on-slurm](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/run-on-slurm)** - 如何在 SLURM 集群上启动分布式 Megatron-LM 训练作业。
|
||
- **[NVIDIA/Megatron-Core/split-pr](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/split-pr)** - 将一个 PR 拆分为多个 PR,以减少所需的 CODEOWNERS 审阅者组数量。
|
||
- **[NVIDIA/Megatron-Core/testing](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/testing)** - Megatron-LM 的测试体系。
|
||
- **[NVIDIA/Megatron-Core/update-golden-values](https://github.com/NVIDIA/skills/tree/main/skills/Megatron-Core/update-golden-values)** - 从 GitHub Actions 工作流运行(仅失败作业或全部作业)刷新 golden values,用平均归一化相对差异为变更打分,并生成可直接用于 PR 的摘要。
|
||
|
||
### Model-Optimizer
|
||
|
||
- **[NVIDIA/Model-Optimizer/accessing-mlflow](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/accessing-mlflow)** - 查询并浏览存储在 MLflow 中的评估结果。
|
||
- **[NVIDIA/Model-Optimizer/debug](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/debug)** - 通过基于文件的命令中继 (tools/debugger) 在远程 Docker 容器内运行命令。
|
||
- **[NVIDIA/Model-Optimizer/deployment](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/deployment)** - 使用 vLLM、SGLang 或 TRT-LLM 将量化或未量化的 LLM checkpoint 部署为 OpenAI 兼容的 API 端点。
|
||
- **[NVIDIA/Model-Optimizer/evaluation](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/evaluation)** - 使用 NeMo Evaluator Launcher (NEL) 评估量化或未量化 LLM 的准确性。
|
||
- **[NVIDIA/Model-Optimizer/launching-evals](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/launching-evals)** - 通过 nemo-evaluator-launcher 运行、监控、分析和调试 LLM 评估。
|
||
- **[NVIDIA/Model-Optimizer/monitor](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/monitor)** - 在 SLURM 集群上监控已提交的作业(PTQ、评估、部署)。
|
||
- **[NVIDIA/Model-Optimizer/ptq](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/ptq)** - 当用户要求「quantize a model」、「run PTQ」、「post-training quantization」、「NVFP4 quantization」、「FP8 quantization」、「INT8 quantization」、「INT4 AW...」时应使用此技能。
|
||
- **[NVIDIA/Model-Optimizer/release-cherry-pick](https://github.com/NVIDIA/skills/tree/main/skills/Model-Optimizer/release-cherry-pick)** - 将已合并且标记为发布分支的 PR cherry-pick 到该分支,然后打开 PR 并应用 cherry-pick-done 标签。
|
||
|
||
### NeMo-Evaluator
|
||
|
||
- **[NVIDIA/NeMo-Evaluator/byob](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Evaluator/byob)** - 使用 BYOB decorator 框架创建自定义 LLM 评估基准。
|
||
|
||
### NeMo-Evaluator-Launcher
|
||
|
||
- **[NVIDIA/NeMo-Evaluator-Launcher/accessing-mlflow](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Evaluator-Launcher/accessing-mlflow)** - 查询并浏览存储在 MLflow 中的评估结果。
|
||
- **[NVIDIA/NeMo-Evaluator-Launcher/launching-evals](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Evaluator-Launcher/launching-evals)** - 通过 nemo-evaluator-launcher 运行、监控、分析和调试 LLM 评估。
|
||
- **[NVIDIA/NeMo-Evaluator-Launcher/nel-assistant](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Evaluator-Launcher/nel-assistant)** - NeMo Evaluator Launcher (NEL) 的交互式配置向导。
|
||
|
||
### NeMo-Gym
|
||
|
||
- **[NVIDIA/NeMo-Gym/add-benchmark](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Gym/add-benchmark)** - > 向 NeMo-Gym 添加新基准或训练环境的指南。
|
||
- **[NVIDIA/NeMo-Gym/nemo-gym-debugging](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Gym/nemo-gym-debugging)** - >- 在调试 Nemo Gym 运行或奖励分析 (reward profiling) 作业时使用。
|
||
- **[NVIDIA/NeMo-Gym/nemo-gym-docs](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Gym/nemo-gym-docs)** - > 维护 NeMo Gym Fern 文档站点——在 fern/ 下添加、更新、移动或删除页面。
|
||
- **[NVIDIA/NeMo-Gym/nemo-gym-pivot-datasets](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Gym/nemo-gym-pivot-datasets)** - >- 在从 rollout、trajectory、chat-completion、Responses API 或 tool-call 制品创建、验证或记录 Nemo Gym pivot 数据集时使用。
|
||
- **[NVIDIA/NeMo-Gym/nemo-gym-reward-profiling](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-Gym/nemo-gym-reward-profiling)** - >- 用于帮助用户上手 Nemo Gym 奖励分析 (reward profiling)。
|
||
|
||
### NeMo-RL
|
||
|
||
- **[NVIDIA/NeMo-RL/auto-research](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/auto-research)** - 面向定向假设检验与开放式探索的自主 NeMo-RL 研究智能体工作流。
|
||
- **[NVIDIA/NeMo-RL/brev-etiquette](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/brev-etiquette)** - 针对在 /home/ubuntu/RL 中工作、工作区磁盘空间有限、拥有更大 /ephemeral 卷以及可选 /home/ubuntu/RL/.env 机密的 NeMo-RL 智能体的 Brev 实例操作指南。
|
||
- **[NVIDIA/NeMo-RL/build-and-dependency](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/build-and-dependency)** - NeMo-RL 的构建与依赖管理。
|
||
- **[NVIDIA/NeMo-RL/cicd](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/cicd)** - NeMo-RL 的 CI/CD 参考指南。
|
||
- **[NVIDIA/NeMo-RL/config-conventions](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/config-conventions)** - NeMo-RL 的配置约定。
|
||
- **[NVIDIA/NeMo-RL/contributing](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/contributing)** - NeMo-RL 的贡献约定。
|
||
- **[NVIDIA/NeMo-RL/copyright](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/copyright)** - NeMo-RL 的 NVIDIA 版权声明头要求。
|
||
- **[NVIDIA/NeMo-RL/docs](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/docs)** - NeMo-RL 的文档约定。
|
||
- **[NVIDIA/NeMo-RL/error-handling](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/error-handling)** - NeMo-RL 的错误处理指南。
|
||
- **[NVIDIA/NeMo-RL/launch-nemo-rl](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/launch-nemo-rl)** - 通过 nrl-k8s CLI 在 Kubernetes 集群上启动、监控、停止和调试 NeMo-RL recipes 的操作手册。
|
||
- **[NVIDIA/NeMo-RL/linting-and-formatting](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/linting-and-formatting)** - NeMo-RL 的代码风格指南(Python 与 shell)。
|
||
- **[NVIDIA/NeMo-RL/review-pr](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/review-pr)** - 针对 NVIDIA-NeMo/RL pull request 的交互式代码审查。
|
||
- **[NVIDIA/NeMo-RL/session-memory](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/session-memory)** - 管理编码智能体的持久工作会话记忆。
|
||
- **[NVIDIA/NeMo-RL/testing](https://github.com/NVIDIA/skills/tree/main/skills/NeMo-RL/testing)** - NeMo-RL 的测试约定。
|
||
|
||
### NemoClaw
|
||
|
||
- **[NVIDIA/NemoClaw/nemoclaw-contributor-create-pr](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-contributor-create-pr)** - 创建遵循 NemoClaw PR 模板的 GitHub pull request。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-contributor-update-docs](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-contributor-update-docs)** - 扫描近期 git 提交,查找影响面向用户行为的变更,然后起草或更新相应文档页面,并刷新生成的用户技能,为发布做准备。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-cross-issue-sweep](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-cross-issue-sweep)** - 扫描其他 open issue,找出给定 PR 可能一并修复或意外破坏的 issue。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-cut-release-tag](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-cut-release-tag)** - 切出新 semver 发布版本——通过 bump-version.ts 提升所有版本字符串,开启发布 PR,合并后为 main 打 tag 并 push。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-day](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-day)** - 运行 NemoClaw 的日间维护者循环,优先处理带有当前版本目标标签的条目。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-evening](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-evening)** - 运行 NemoClaw 的日终维护者交接流程。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-find-review-pr](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-find-review-pr)** - 查找带有 security 和 priority-high 标签的 open GitHub PR,将每个 PR 关联到对应 issue,检测重复项(多个 PR 修复同一 issue),并呈现审查候选表格……
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-morning](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-morning)** - 运行 NemoClaw 的早间维护者站会。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-normalize-title-tags](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-normalize-title-tags)** - 规范化 GitHub issue 和 PR 标题,不区分大小写地移除任何方括号包裹的 [NemoClaw] 标签,即使标签出现在标题后部。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-pr-comparator](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-pr-comparator)** - 比较针对同一 issue 的竞争性 PR,并推荐应合并哪一个。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-security-code-review](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-security-code-review)** - 对 GitHub PR 或 issue 中的代码变更进行全面安全审查。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-maintainer-triage](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-maintainer-triage)** - 针对 NVIDIA/NemoClaw issue 和 PR 的 AI 辅助标签分流。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-skills-guide](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-skills-guide)** - 从这里开始。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-agent-skills](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-agent-skills)** - 介绍 NemoClaw 内置的 agent skills,以及如何通过克隆仓库访问它们。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-configure-inference](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-configure-inference)** - 将 NemoClaw 连接到本地推理服务器。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-configure-security](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-configure-security)** - 为 NemoClaw 中每个可配置的安全控制项提供风险框架。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-deploy-remote](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-deploy-remote)** - 说明如何在远程 GPU 实例上运行 NemoClaw,包括已弃用的 Brev 兼容路径,以及首选的安装器加 onboard 流程。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-get-started](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-get-started)** - 安装 NemoClaw,启动 sandbox,并运行首个 agent prompt。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-manage-policy](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-manage-policy)** - 在 sandbox 策略中添加、移除或修改允许的 endpoint。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-manage-sandboxes](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-manage-sandboxes)** - 说明 quickstart 之后的运维任务:列出 sandbox、状态与健康检查、日志、诊断、端口转发、多个 sandbox、凭据重置、重建、网络……
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-monitor-sandbox](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-monitor-sandbox)** - 检查 sandbox 健康状态,追踪 agent 行为,并诊断问题。
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-overview](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-overview)** - 说明 OpenClaw、OpenShell 和 NemoClaw 如何构成生态系统、NemoClaw 在技术栈中的位置、NemoClaw 相较社区 sandbox 的额外能力,以及何时应优先选择 NemoClaw……
|
||
- **[NVIDIA/NemoClaw/nemoclaw-user-reference](https://github.com/NVIDIA/skills/tree/main/skills/NemoClaw/nemoclaw-user-reference)** - 介绍 NemoClaw 的 plugin 与 blueprint 架构,以及它们如何编排 OpenClaw sandbox。
|
||
|
||
### TensorRT-LLM
|
||
|
||
- **[NVIDIA/TensorRT-LLM/ad-accuracy-debug](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-accuracy-debug)** - > 调试 AutoDeploy 相对参考分数(PyTorch backend 或已发布 baseline)的精度回归。
|
||
- **[NVIDIA/TensorRT-LLM/ad-add-fusion-transformation](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-add-fusion-transformation)** - > Claude Code skill(trtllm-agent-toolkit):在 TensorRT-LLM checkout 的 transform/library/ 下实现或扩展 TensorRT-LLM AutoDeploy fusion transform。
|
||
- **[NVIDIA/TensorRT-LLM/ad-conf-check](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-conf-check)** - > 通过分析服务器日志,并可选择性地分析 graph dump(AD_DUMP_GRAPHS_DIR),检查 AutoDeploy YAML 配置是否已实际应用。
|
||
- **[NVIDIA/TensorRT-LLM/ad-graph-dump](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-graph-dump)** - > 通过 AD_DUMP_GRAPHS_DIR 启用并解读 TensorRT-LLM AutoDeploy FX graph 文本 dump。
|
||
- **[NVIDIA/TensorRT-LLM/ad-layer-visualizer](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-layer-visualizer)** - > 将 AutoDeploy FX graph 文本 dump 中的特定 transformer decoder layer 可视化为分层 DOT/PNG 图。
|
||
- **[NVIDIA/TensorRT-LLM/ad-model-onboard](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/ad-model-onboard)** - > 使用参考 custom ops 将 HuggingFace 模型转换为仅 prefill 的 AutoDeploy custom model,并通过分层等价性测试进行验证。
|
||
- **[NVIDIA/TensorRT-LLM/exec-local-compile](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/exec-local-compile)** - 在 Docker 容器内的计算节点上编译 TensorRT-LLM。
|
||
- **[NVIDIA/TensorRT-LLM/exec-slurm-compile](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/exec-slurm-compile)** - 在 SLURM 集群上编译 TensorRT-LLM。
|
||
- **[NVIDIA/TensorRT-LLM/kernel-cute-writing](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/kernel-cute-writing)** - > 使用 NVIDIA CuTe DSL(CUTLASS 4.x Python API)编写并实现 GPU kernel——不适用于 Triton、CUDA C++ 或概念性说明。
|
||
- **[NVIDIA/TensorRT-LLM/kernel-tileir-optimization](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/kernel-tileir-optimization)** - > 针对 Blackwell GPU(sm_100+)上的 NVIDIA TileIR backend 优化现有 Triton kernel。
|
||
- **[NVIDIA/TensorRT-LLM/kernel-triton-writing](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/kernel-triton-writing)** - > 仅用于 OpenAI Triton(@triton.jit)kernel 开发。
|
||
- **[NVIDIA/TensorRT-LLM/perf-analysis](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-analysis)** - > 性能分析协调工作流。
|
||
- **[NVIDIA/TensorRT-LLM/perf-host-analysis](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-host-analysis)** - > 从 nsys trace 分析 TensorRT-LLM 推理中的 host/CPU 开销。
|
||
- **[NVIDIA/TensorRT-LLM/perf-host-optimization](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-host-optimization)** - 使用 line_profiler 分析并优化 TensorRT-LLM 的 host/CPU 开销(计划支持 nsys)。
|
||
- **[NVIDIA/TensorRT-LLM/perf-nsight-compute-analysis](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-nsight-compute-analysis)** - > 分析 ncu(NVIDIA Nsight Compute)profiling 输出:SOL% 瓶颈分类、roofline 分析、occupancy 诊断、内存层次分析、warp stall 分析、指标……
|
||
- **[NVIDIA/TensorRT-LLM/perf-nsight-systems](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-nsight-systems)** - >- 用于系统级时间线 profiling 的 Nsight Systems(nsys)CLI。
|
||
- **[NVIDIA/TensorRT-LLM/perf-optimization](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-optimization)** - > 性能优化协调 playbook。
|
||
- **[NVIDIA/TensorRT-LLM/perf-torch-cuda-graphs](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-torch-cuda-graphs)** - >- 将 CUDA Graphs 应用于 PyTorch 工作负载——API 选择(torch.compile、PyTorch make_graphed_callables、TE make_graphed_callables、MCore CudaGraphManager、FullCudaGraphWrapper、m……
|
||
- **[NVIDIA/TensorRT-LLM/perf-torch-sync-free](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-torch-sync-free)** - >- 识别并消除 PyTorch 代码中的 host-device 同步。
|
||
- **[NVIDIA/TensorRT-LLM/perf-workload-profiling](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/perf-workload-profiling)** - > 用于为工作负载计时的代码插桩。
|
||
- **[NVIDIA/TensorRT-LLM/trtllm-code-contribution](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/trtllm-code-contribution)** - > 向 TensorRT-LLM 贡献代码的最佳实践。
|
||
- **[NVIDIA/TensorRT-LLM/trtllm-codebase-exploration](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/trtllm-codebase-exploration)** - > 在实现新功能或优化之前系统性探索 TensorRT-LLM 代码库的方法。
|
||
- **[NVIDIA/TensorRT-LLM/trtllm-flashinfer-upgrade](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/trtllm-flashinfer-upgrade)** - >- 升级 TensorRT-LLM 中的 flashinfer-python 版本。
|
||
- **[NVIDIA/TensorRT-LLM/trtllm-moe-develop](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/trtllm-moe-develop)** - >- 审查、设计并重构 TensorRT-LLM PyTorch MoE 代码,以契合架构、保持代码整洁,并提升可维护性与可测试性。
|
||
- **[NVIDIA/TensorRT-LLM/trtllm-serve-config-guide](https://github.com/NVIDIA/skills/tree/main/skills/TensorRT-LLM/trtllm-serve-config-guide)** - 生成面向基础 aggregate 单节点 PyTorch serving 的、有据可查的初始 `trtllm-serve --config` YAML,并与入库的 TensorRT-LLM 配置及部署文档保持一致。
|
||
|
||
### TileGym
|
||
|
||
- **[NVIDIA/TileGym/adding-cutile-kernel](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/adding-cutile-kernel)** - 向 TileGym 添加新的 cuTile GPU 内核算子。
|
||
- **[NVIDIA/TileGym/converting-cutile-to-julia](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/converting-cutile-to-julia)** - 将 cuTile Python GPU 内核(@ct.kernel)转换为 cuTile.jl Julia 等价实现。
|
||
- **[NVIDIA/TileGym/converting-cutile-to-triton](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/converting-cutile-to-triton)** - 将 cuTile GPU 内核(@ct.kernel)转换为 Triton(@triton.jit)。
|
||
- **[NVIDIA/TileGym/cutile-autotuning](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/cutile-autotuning)** - 在添加、修改、优化或调试 CuTile 自动调优代码时使用。
|
||
- **[NVIDIA/TileGym/cutile-python](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/cutile-python)** - cuTile 编程专家助手。
|
||
- **[NVIDIA/TileGym/improve-cutile-kernel-perf](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/improve-cutile-kernel-perf)** - 通过系统化性能剖析、瓶颈分析、IR 对比与针对性调优,迭代优化 cuTile 内核性能。
|
||
- **[NVIDIA/TileGym/monkey-patch-kernels-to-transformers](https://github.com/NVIDIA/skills/tree/main/skills/TileGym/monkey-patch-kernels-to-transformers)** - 通过替换库的 submodule(s) 以及某些 class(es) 的实现,并对某些 class(es) 进行 in... 补丁,将 TileGym 内核集成到 Hugging Face `transformers` 模型中。
|
||
|
||
### cuopt
|
||
|
||
- **[NVIDIA/cuopt/cuopt-developer](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-developer)** - 修改、构建、测试、调试并为 NVIDIA cuOpt(C++/CUDA、Python、server、CI)做贡献。
|
||
- **[NVIDIA/cuopt/cuopt-install](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-install)** - 为 Python、C 或作为 server 安装 cuOpt(pip、conda、Docker)——系统要求、安装命令与验证。
|
||
- **[NVIDIA/cuopt/cuopt-numerical-optimization-api-c](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-numerical-optimization-api-c)** - 使用 cuOpt 求解 LP、MILP 与 QP(beta)——仅 C API。
|
||
- **[NVIDIA/cuopt/cuopt-numerical-optimization-api-cli](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-numerical-optimization-api-cli)** - 使用 cuOpt 求解 LP、MILP 与 QP(beta)——仅 CLI(MPS 文件、cuopt_cli)。
|
||
- **[NVIDIA/cuopt/cuopt-numerical-optimization-api-python](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-numerical-optimization-api-python)** - 使用 Python API 求解线性规划(Linear Programming, LP)、混合整数线性规划(Mixed-Integer Linear Programming, MILP)与二次规划(Quadratic Programming, QP,beta)。
|
||
- **[NVIDIA/cuopt/cuopt-routing-api-python](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-routing-api-python)** - 使用 cuOpt 进行车辆路径规划(Vehicle routing, VRP、TSP、PDP)——仅 Python API。
|
||
- **[NVIDIA/cuopt/cuopt-server-api-python](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-server-api-python)** - cuOpt REST server — 启动服务器、端点、Python/curl 客户端示例。
|
||
- **[NVIDIA/cuopt/cuopt-server-common](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-server-common)** - cuOpt REST server — 功能说明与请求流转方式。
|
||
- **[NVIDIA/cuopt/cuopt-user-rules](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/cuopt-user-rules)** - 终端用户调用 NVIDIA cuOpt(routing/LP/MILP/QP/install/server)的基础规则。
|
||
- **[NVIDIA/cuopt/numerical-optimization-formulation](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/numerical-optimization-formulation)** - 数值优化(LP、MILP、QP)——概念、问题文本解析与建模模式。
|
||
- **[NVIDIA/cuopt/routing-formulation](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/routing-formulation)** - 车辆路径规划(VRP、TSP、PDP)——问题类型与数据要求。
|
||
- **[NVIDIA/cuopt/skill-evolution](https://github.com/NVIDIA/skills/tree/main/skills/cuopt/skill-evolution)** - 在解决非平凡问题后,识别可泛化的经验并提议 skill 更新,使后续交互自动受益。
|
||
|
||
### deepstream
|
||
|
||
- **[NVIDIA/deepstream/deepstream-dev](https://github.com/NVIDIA/skills/tree/main/skills/deepstream/deepstream-dev)** - 使用 Python pyservicemaker API 进行 NVIDIA DeepStream SDK 9.0 开发。
|
||
- **[NVIDIA/deepstream/deepstream-import-vision-model](https://github.com/NVIDIA/skills/tree/main/skills/deepstream/deepstream-import-vision-model)** - > 使用此 skill 将 HuggingFace 或 NVIDIA NGC 上的任意视觉模型以端到端自动化方式接入 NVIDIA DeepStream 流水线:ONNX 下载、SafeTensors 导出、TRT engi...
|
||
|
||
### nemotron-voice-agent
|
||
|
||
- **[NVIDIA/nemotron-voice-agent/nemotron-voice-agent-deploy](https://github.com/NVIDIA/skills/tree/main/skills/nemotron-voice-agent/nemotron-voice-agent-deploy)** - 在 Workstation(x86)、Jetson Thor 或 Cloud NIMs 上部署 Nemotron Voice Agent。
|
||
|
||
### rag
|
||
|
||
- **[NVIDIA/rag/rag-blueprint](https://github.com/NVIDIA/skills/tree/main/skills/rag/rag-blueprint)** - "NVIDIA RAG Blueprint — 部署、配置、故障排查与管理。
|
||
|
||
### video-search-and-summarization
|
||
|
||
- **[NVIDIA/video-search-and-summarization/alerts](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/alerts)** - 在 alerts profile 部署后管理与监控 VSS 告警。
|
||
- **[NVIDIA/video-search-and-summarization/deploy](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/deploy)** - 使用以 compose 为中心的工作流部署、调试或拆除任意 VSS profile — 带环境变量覆盖的 config(dry-run)、审阅解析后的 compose,然后 compose up。
|
||
- **[NVIDIA/video-search-and-summarization/report](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/report)** - 通过发现已部署的 VSS agent、查询 POST /generate 获取带时间戳的字幕摘要,再将 agent 回复格式化为 th...,生成视频分析报告。
|
||
- **[NVIDIA/video-search-and-summarization/rt-vlm](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/rt-vlm)** - > 在 VSS 3.1 上使用 RTVI VLM 或 RT-VLM 微服务 API 时使用此 skill。
|
||
- **[NVIDIA/video-search-and-summarization/video-analytics](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/video-analytics)** - 通过 VA-MCP server(端口 9901)从 Elastic search 查询视频分析数据与指标。
|
||
- **[NVIDIA/video-search-and-summarization/video-search](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/video-search)** - 使用自然语言搜索视频存档 — 通过融合搜索(Cosmos Embed1 语义搜索 + CV attribute sea...)在录制视频中查找事件、物体、动作与人物。
|
||
- **[NVIDIA/video-search-and-summarization/video-summarization](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/video-summarization)** - 直接调用 VLM NIM 或 Long Video Summarization(LVS)微服务对视频进行摘要。
|
||
- **[NVIDIA/video-search-and-summarization/video-understanding](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/video-understanding)** - 调用 vss agent 对视频执行视频理解,以回答文本问题。
|
||
- **[NVIDIA/video-search-and-summarization/vios](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/vios)** - 查询 VIOS REST API:传感器列表、录制时间线、视频片段提取、快照捕获、添加/删除传感器与流
|
||
- **[NVIDIA/video-search-and-summarization/vss-frag](https://github.com/NVIDIA/skills/tree/main/skills/video-search-and-summarization/vss-frag)** - 使用 VSS video_search_frag 扩展,结合 Long Video Summarization(LVS)、Enterprise RAG 知识检索与人机协同(human-in-the-loop)参数 co...,生成视频摘要报告。
|
||
|
||
</details>
|
||
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by Google Cloud</h3></summary>
|
||
|
||
Google Cloud 官方技能,涵盖 Firebase、BigQuery、Cloud Run、GKE、AlloyDB、Cloud SQL、Gemini Enterprise Agent Platform、网络可观测性以及 Well-Architected Framework。共 19 项技能。
|
||
|
||
- **[google/cloud/agent-platform-skill-registry](https://github.com/google/skills/tree/main/skills/cloud/agent-platform-skill-registry)** - 与 Gemini Enterprise Agent Platform Skill Registry 交互,以创建和搜索可用技能。
|
||
- **[google/cloud/alloydb-basics](https://github.com/google/skills/tree/main/skills/cloud/alloydb-basics)** - 管理 AlloyDB for PostgreSQL 的集群、实例和备份,并与 AlloyDB 模型上下文协议(MCP)工具集成,实现自动化数据库操作。
|
||
- **[google/cloud/bigquery-basics](https://github.com/google/skills/tree/main/skills/cloud/bigquery-basics)** - 管理 BigQuery 中的数据集、表和作业,并与 BigQuery ML 和 Gemini 集成,用于高级数据分析和 AI 驱动的洞察。
|
||
- **[google/cloud/cloud-run-basics](https://github.com/google/skills/tree/main/skills/cloud/cloud-run-basics)** - 管理 Cloud Run 服务、作业和工作池。
|
||
- **[google/cloud/cloud-sql-basics](https://github.com/google/skills/tree/main/skills/cloud/cloud-sql-basics)** - 该文件可生成或说明 Cloud SQL 资源。
|
||
- **[google/cloud/firebase-basics](https://github.com/google/skills/tree/main/skills/cloud/firebase-basics)** - 在从事使用 Firebase 产品或服务的项目时(尤其是移动或 Web 应用)可使用此技能。
|
||
- **[google/cloud/gemini-agents-api](https://github.com/google/skills/tree/main/skills/cloud/gemini-agents-api)** - 在 Gemini Enterprise Agent Platform 上管理自定义 Agent 资源。
|
||
- **[google/cloud/gemini-api](https://github.com/google/skills/tree/main/skills/cloud/gemini-api)** - 指导在 Agent Platform 上通过 Google Gen AI SDK 使用 Gemini API。
|
||
- **[google/cloud/gemini-interactions-api](https://github.com/google/skills/tree/main/skills/cloud/gemini-interactions-api)** - 指导在 Gemini Enterprise Agent Platform 上使用 Gemini Interactions API。
|
||
- **[google/cloud/gke-basics](https://github.com/google/skills/tree/main/skills/cloud/gke-basics)** - 使用黄金路径 Autopilot 配置来规划、创建和配置可用于生产的 Google Kubernetes Engine(GKE)集群。
|
||
- **[google/cloud/google-cloud-networking-observability](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-networking-observability)** - 通过分析日志、指标和诊断信息来排查 Google Cloud 网络问题。
|
||
- **[google/cloud/google-cloud-recipe-auth](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-recipe-auth)** - 为 Google Cloud 服务和 API 的身份验证与授权提供专家指导,涵盖人类用户、服务身份、Application Default Credentials(ADC)以及 b...
|
||
- **[google/cloud/google-cloud-recipe-onboarding](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-recipe-onboarding)** - 为开发者在 Google Cloud 上的入门步骤提供指导,涵盖账户创建、计费设置、项目管理以及部署首个资源。
|
||
- **[google/cloud/google-cloud-waf-cost-optimization](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-cost-optimization)** - 基于 Google Cloud Well-Architected Framework(WAF)为 Google Cloud 工作负载生成成本优化指导。
|
||
- **[google/cloud/google-cloud-waf-operational-excellence](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-operational-excellence)** - 基于 Google Cloud Well-Ar... 中卓越运营支柱的设计原则与建议,为 Google Cloud 工作负载生成面向运维的指导。
|
||
- **[google/cloud/google-cloud-waf-performance-optimization](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-performance-optimization)** - 基于 Google Cloud Well... 中性能优化支柱的设计原则与建议,为 Google Cloud 工作负载生成面向性能的指导。
|
||
- **[google/cloud/google-cloud-waf-reliability](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-reliability)** - 基于 Google Cloud Well-Architected Framework 的设计原则与建议,为 Google Cloud 工作负载生成面向可靠性的指导。
|
||
- **[google/cloud/google-cloud-waf-security](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-security)** - 基于 Google Cloud Well-Architected Framework(WAF)的设计原则与建议,为 Google Cloud 工作负载生成面向安全的指导。
|
||
- **[google/cloud/google-cloud-waf-sustainability](https://github.com/google/skills/tree/main/skills/cloud/google-cloud-waf-sustainability)** - 基于 Google Cloud Well-Architected Framework(WAF)的设计原则与建议,为 Google Cloud 工作负载生成面向可持续性的指导。
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by RedHat</h3></summary>
|
||
|
||
借助由 Red Hat 订阅支持的精选技能、智能体(agents)和 MCP 服务器库,将 AI 能力扩展到整个组织。无论你是正在 Cursor 中优化工作流的 SRE,还是正在构建智能界面的架构师,都可借助可信构建模块,自信地部署和扩展智能体自动化。
|
||
|
||
- **[redhat/cve-skillpack](https://catalog.redhat.com/en/ai/skills/detail/agentic-skill-pack-for-red-hat-customers#agent-and-skills)** - 了解 CVE、检查产品生命周期状态、收集诊断信息,并以合适的严重程度提交支持案例——日常运维必备的 Red Hat 技能。
|
||
|
||
- **[redhat/sre-skillpack](https://catalog.redhat.com/en/ai/skills/detail/agentic-skill-pack-for-site-reliability-engineers)** - 在整个 RHEL 集群中发现、修复并验证 CVE——通过单一工作流编排 Red Hat Lightspeed 和 Ansible Automation Platform。
|
||
|
||
- **[redhat/openshift-skillpack](https://catalog.redhat.com/en/ai/skills/detail/agentic-skill-pack-for-red-hat-openshift)** - 通过单一对话式工作流,对 OpenShift 集群进行配置、清点与报告——覆盖 Assisted Installer、OCM、ROSA、ARO 及 kubeconfig 集群。
|
||
|
||
- **[redhat/openshift-virtualization](https://catalog.redhat.com/en/ai/skills/detail/agentic-skill-pack-for-red-hat-openshift-virtualization)** - 通过单一对话式工作流,在 OpenShift Virtualization 上管理完整的 VM 生命周期——创建、克隆、快照、恢复、重新平衡与报告。
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Skills by Cypress</h3></summary>
|
||
|
||
Cypress 官方发布的技能,用于帮助创建、维护、理解并修复 Cypress 测试。共 3 项技能。
|
||
|
||
- **[cypress-io/cypress-author](https://github.com/cypress-io/ai-toolkit/tree/main/skills/cypress-author)** - 创建、更新并修复 Cypress E2E 与组件测试。
|
||
- **[cypress-io/cypress-explain](https://github.com/cypress-io/ai-toolkit/tree/main/skills/cypress-explain)** - 解释 Cypress E2E 与组件测试,并解答有关 Cypress 用法与行为的问题。
|
||
- **[cypress-io/cypress-docs](https://github.com/cypress-io/ai-toolkit/tree/main/skills/cypress-docs)** - 从官方文档中搜索并提取 Cypress 相关信息。
|
||
|
||
</details>
|
||
|
||
|
||
### Community Skills
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Vector Databases</h3></summary>
|
||
|
||
- **[qdrant/skills](https://github.com/qdrant/skills)** - 面向 Qdrant 向量搜索的智能体技能,涵盖扩展、性能优化、搜索质量、监控、部署、模型迁移、版本升级,以及在 Python、TypeScript、Rust、Go、.NET 和 Java 中的 SDK 使用。
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">Marketing</h3></summary>
|
||
|
||
- **[BrianRWagner/ai-marketing-skills](https://github.com/BrianRWagner/ai-marketing-skills)** - 17 个营销框架,涵盖冷启动外联、首页审计、社交卡片等
|
||
- **[AgriciDaniel/claude-seo](https://github.com/AgriciDaniel/claude-seo)** - 通用 SEO 技能,用于全面的网站分析与优化
|
||
- **[wshuyi/x-article-publisher-skill](https://github.com/wshuyi/x-article-publisher-skill)** - 将文章发布到 X/Twitter
|
||
- **[CosmoBlk/email-marketing-bible](https://github.com/CosmoBlk/email-marketing-bible)** - 5.5 万字电子邮件营销指南,以 AI 技能形式呈现
|
||
- **[smixs/creative-director-skill](https://github.com/smixs/creative-director-skill)** - 具备递归自评能力的 AI 创意总监:20+ 种方法论(SIT、TRIZ、Bisociation、SCAMPER、Synectics),参照 Cannes/D&AD/HumanKind 校准的三轴评估,从简报到演示的 5 阶段流程
|
||
- **[Xquik-dev/x-twitter-scraper](https://github.com/Xquik-dev/x-twitter-scraper)** - 推文搜索、个人资料推文、粉丝导出、媒体、发帖、回复、MCP
|
||
- **[Xquik-dev/tweetclaw](https://github.com/Xquik-dev/tweetclaw)** - 发布推文、回复、私信;搜索、监控、举办抽奖活动
|
||
- **[SHADOWPR0/beautiful_prose](https://github.com/SHADOWPR0/beautiful_prose)** - 为历久弥新、有力度的英文散文提供硬朗写作风格契约,避免 AI 写作痕迹
|
||
- **[blader/humanizer](https://github.com/blader/humanizer)** - 去除文本中 AI 生成写作的痕迹,使其听起来更自然、更有人味
|
||
- **[MohamedAbdallah-14/unslop](https://github.com/MohamedAbdallah-14/unslop)** - 移除具名 AI 写作特征(三连修辞、破折号堆砌、含糊回避叠层、谄媚式开篇、如 "delve"/"crucial" 等套话词汇)。提供 lint/rewrite 分离模式,便于审计自有文本而不自动改写。五个强度级别,MIT
|
||
- **[Eronred/aso-skills](https://github.com/Eronred/aso-skills)** - 30+ 个 App Store Optimization(应用商店优化)技能,涵盖关键词研究、元数据优化、竞品分析、创意优化,以及通过 Appeeky API 的移动增长策略
|
||
- **[degausai/wonda](https://github.com/degausai/wonda)** - AI 内容创作:图像、视频、音乐、音频、编辑、发布
|
||
- **[gitroomhq/postiz-agent](https://github.com/gitroomhq/postiz-agent)** - 以编程方式在 28+ 个平台上安排社交媒体帖子
|
||
- **[taisly/agent](https://github.com/taisly/agent)** - Codex 插件、Agent Skill、CLI 和 MCP 服务器,通过 Taisly 将已审核的短视频发布到 TikTok、Instagram Reels、YouTube Shorts、X 和 Facebook
|
||
- **[indranilbanerjee/digital-marketing-pro](https://github.com/indranilbanerjee/digital-marketing-pro)** - 150 技能参与方法论——12 部分策略流程、25 个专业代理、符合欧盟 AI 法案第 50 条(C2PA 签名)、涵盖 Google AI Mode 的 6 平台 AEO/GEO
|
||
- **[infrasity-labs/dev-gtm-claude-skills](https://github.com/infrasity-labs/dev-gtm-claude-skills)**: GTM(go-to-market,进入市场)导向的技能集合,面向开发者进入市场工作流,涵盖发布规划、定位与外联序列。
|
||
- **[nowork-studio/NotFair](https://github.com/nowork-studio/NotFair)** - 基于实时数据的 SEO、GEO、Google Ads 和 Meta Ads 技能
|
||
- **[aaron-he-zhu/aaron-marketing-skills](https://github.com/aaron-he-zhu/aaron-marketing-skills)** - 69 个营销技能,统一契约覆盖 SEO/GEO、网红营销、付费广告与电子邮件,配备 5 个基准驱动的审计关卡(CORE-EEAT、CITE、C³、ROAS、SEND)及免密钥数据连接器
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">生产力与协作</h3></summary>
|
||
|
||
- **[PSPDFKit-labs/nutrient-agent-skill](https://github.com/PSPDFKit-labs/nutrient-agent-skill)** - 通过 Nutrient DWS API 进行文档处理:转换(PDF/DOCX/XLSX/PPTX/HTML/图像)、提取文本/表格、OCR(20+ 种语言)、脱敏 PII(模式 + AI)、水印、数字签名、表单填写。[MCP server](https://www.npmjs.com/package/@nutrient-sdk/dws-mcp-server) also available.
|
||
- **[notiondevs/Notion Skills for Claude](https://www.notion.so/notiondevs/Notion-Skills-for-Claude-28da4445d27180c7af1df7d8615723d0)** - 用于与 Notion 协作的技能
|
||
- **[op7418/NanoBanana-PPT-Skills](https://github.com/op7418/NanoBanana-PPT-Skills)** - AI 驱动的 PPT 生成,支持文档分析与风格化图像
|
||
- **[zarazhangrui/frontend-slides](https://github.com/zarazhangrui/frontend-slides)** - 生成动画丰富的 HTML 演示文稿,并提供视觉风格预览
|
||
- **[gokapso/integrate-whatsapp](https://github.com/gokapso/agent-skills/tree/master/skills/integrate-whatsapp)** - 连接 WhatsApp、设置 webhook 并发送消息
|
||
- **[gokapso/automate-whatsapp](https://github.com/gokapso/agent-skills/tree/master/skills/automate-whatsapp)** - 通过工作流与代理构建 WhatsApp 自动化
|
||
- **[gokapso/observe-whatsapp](https://github.com/gokapso/agent-skills/tree/master/skills/observe-whatsapp)** - 调试 WhatsApp 投递问题并运行健康检查
|
||
- **[PleasePrompto/notebooklm-skill](https://github.com/PleasePrompto/notebooklm-skill)** - 与 NotebookLM 交互,进行基于文档的对话
|
||
- **[obra/superpowers-lab](https://github.com/obra/superpowers-lab)** - Claude superpowers 的实验环境
|
||
- **[obra/brainstorming](https://github.com/obra/superpowers/blob/main/skills/brainstorming/SKILL.md)** - 生成并探索创意
|
||
- **[obra/writing-plans](https://github.com/obra/superpowers/blob/main/skills/writing-plans/SKILL.md)** - 撰写战略文档
|
||
- **[obra/executing-plans](https://github.com/obra/superpowers/blob/main/skills/executing-plans/SKILL.md)** - 实施并运行战略计划
|
||
- **[obra/dispatching-parallel-agents](https://github.com/obra/superpowers/blob/main/skills/dispatching-parallel-agents/SKILL.md)** - 协调多个并行代理
|
||
- **[obra/sharing-skills](https://github.com/obra/superpowers/blob/main/skills/sharing-skills/SKILL.md)** - 分发并传达能力
|
||
- **[obra/using-superpowers](https://github.com/obra/superpowers/blob/main/skills/using-superpowers/SKILL.md)** - 运用核心平台能力
|
||
- **[op7418/Youtube-clipper-skill](https://github.com/op7418/Youtube-clipper-skill)** - YouTube 片段生成与编辑,配备自动化工作流
|
||
- **[ognjengt/founder-skills](https://github.com/ognjengt/founder-skills)** - 面向创始人的 Claude 技能,内置打包好的创业工作流
|
||
- **[EveryInc/charlie-cfo-skill](https://github.com/EveryInc/charlie-cfo-skill)** - 受 Charlie Munger 启发的自力更生型 CFO 财务管理
|
||
- **[openaccountants/openaccountants](https://github.com/openaccountants/openaccountants)** - 覆盖 134 个国家的 371 项税务分类技能
|
||
- **[wrsmith108/linear-claude-skill](https://github.com/wrsmith108/linear-claude-skill)** - 管理 Linear 议题、项目与团队
|
||
- **[Shpigford/readme](https://github.com/Shpigford/skills/tree/main/readme)** - 生成全面的项目文档
|
||
- **[hanfang/claude-memory-skill](https://github.com/hanfang/claude-memory-skill)** - 极简、低摩擦的分层记忆系统,配备后台代理与基于文件系统的持久化
|
||
- **[kreuzberg-dev/kreuzberg](https://github.com/kreuzberg-dev/kreuzberg/tree/main/skills/kreuzberg)** - 从 62+ 种文档格式中提取文本、表格与元数据
|
||
- **[Paramchoudhary/ResumeSkills](https://github.com/Paramchoudhary/ResumeSkills)** - 20 项专业技能,涵盖简历优化、ATS 分析、面试准备与职业转型
|
||
- **[RoundTable02/tutor-skills](https://github.com/RoundTable02/tutor-skills)** - 将文档或代码库转换为带互动测验的 Obsidian StudyVault
|
||
- **[NeoLabHQ/write-concisely](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/docs/skills/write-concisely)** - 运用经典著作 *The Elements of Style* 的原则,通过消除赘语并改进结构,使文档与写作更清晰、更专业
|
||
- **[ReScienceLab/opc-skills](https://github.com/ReScienceLab/opc-skills)** - 面向个体创业者的 Agent 技能,集成 SEO、geo 与 LLM 工具
|
||
- **[SeanZoR/claude-speed-reader](https://github.com/SeanZoR/claude-speed-reader)** - 以 600+ WPM 速读 Claude 回复,采用 RSVP 与 Spritz 风格 ORP 高亮
|
||
- **[Charlie85270/Dorothy](https://github.com/Charlie85270/Dorothy)** - 通过自动化与 MCP 服务器编排多个 AI CLI 代理
|
||
- **[Digidai/product-manager-skills](https://github.com/Digidai/product-manager-skills)** - 高级 PM 代理,内置 30+ 框架与 SaaS 指标
|
||
- **[deusyu/translate-book](https://github.com/deusyu/translate-book)** - 通过可恢复的并行子代理翻译图书(PDF/DOCX/EPUB)
|
||
- **[mvanhorn/last30days-skill](https://github.com/mvanhorn/last30days-skill)** - 跨 Reddit、X、YouTube、HN、Polymarket 与全网研究任意主题,按点赞、喜欢与真金白银排序,而非编辑精选
|
||
- **[santifer/career-ops](https://github.com/santifer/career-ops)** - 14 项技能的 AI 求职套件:A-F 评分的 JD 评估、ATS 优化 PDF 生成、门户扫描器(Greenhouse/Ashby/Lever)、STAR+R 面试准备、批处理,以及 Go 仪表盘 TUI
|
||
- **[Linked-API/linkedin](https://github.com/Linked-API/linkedin-skills/tree/main/linkedin)** - 从 Claude Code、Codex、Cursor 和 Windsurf 获取 LinkedIn 个人资料、搜索人与公司、发送消息、管理连接、创建帖子、点赞、评论,并运行自定义 LinkedIn 工作流
|
||
- **[pattern-ai-labs/agentcall](https://github.com/pattern-ai-labs/agentcall)** - 让你的 AI 代理加入 Google Meet、Zoom、Teams 通话,像真实队友一样协作
|
||
- **[Sendmux/skills](https://github.com/Sendmux/skills)** - 面向代理的 Sendmux 电子邮件与邮箱工作流
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">开发与测试</h3></summary>
|
||
|
||
- **[robzolkos/skill-rails-upgrade](https://github.com/robzolkos/skill-rails-upgrade)** - 分析 Rails 应用并提供升级评估
|
||
- **[Shpigford/screenshots](https://github.com/Shpigford/skills/tree/main/screenshots)** - 使用 Playwright 生成营销截图
|
||
- **[antonbabenko/terraform-skill](https://github.com/antonbabenko/terraform-skill)** - Terraform 与 OpenTofu 模式:测试、模块、状态、CI/CD。
|
||
- **[zxkane/aws-skills](https://github.com/zxkane/aws-skills)** - 面向基础设施自动化与云架构模式的 AWS 开发
|
||
- **[Rootly-AI-Labs/rootly-incident-responder](https://github.com/Rootly-AI-Labs/Rootly-MCP-server/blob/main/examples/skills/rootly-incident-responder.md)** - AI 驱动的事件响应,具备 ML 相似度匹配、解决方案建议与值班协调。需要 [Rootly MCP Server](https://github.com/Rootly-AI-Labs/Rootly-MCP-server)
|
||
- **[conorluddy/ios-simulator-skill](https://github.com/conorluddy/ios-simulator-skill)** - 控制 iOS Simulator
|
||
- **[ramzesenok/iOS-Accessibility-Audit-Skill](https://github.com/ramzesenok/iOS-Accessibility-Audit-Skill)** - 根据无障碍规范审计 iOS 应用
|
||
- **[truongduy2611/app-store-preflight-skills](https://github.com/truongduy2611/app-store-preflight-skills)** - 扫描 iOS/macOS 项目,在提交前捕获易导致 App Store 拒审的常见错误
|
||
- **[coderabbitai/skills](https://github.com/coderabbitai/skills)** - 面向编码代理的代码审查与 PR 自动修复工作流
|
||
- **[sanjay3290/postgres](https://github.com/sanjay3290/ai-skills/tree/main/skills/postgres)** - 对 PostgreSQL 数据库执行安全的只读 SQL 查询
|
||
- **[sanjay3290/deep-research](https://github.com/sanjay3290/ai-skills/tree/main/skills/deep-research)** - 使用 Gemini Deep Research Agent 进行自主多步骤研究
|
||
- **[jthack/ffuf-claude-skill](https://github.com/jthack/ffuf_claude_skill)** - 使用 ffuf 进行 Web 模糊测试
|
||
- **[lackeyjb/playwright-skill](https://github.com/lackeyjb/playwright-skill)** - 使用 Playwright 进行浏览器自动化
|
||
- **[ibelick/ui-skills](https://github.com/ibelick/ui-skills)** - 带有明确主张且持续演进的约束,用于指导代理构建界面
|
||
- **[muthuishere/hand-drawn-diagrams](https://github.com/muthuishere/hand-drawn-diagrams)** - 根据提示生成手绘风格 Excalidraw 图表——动画 SVG、托管编辑链接与 PNG 导出。兼容 Claude Code、Codex、Gemini CLI 及任何支持标准 skill 路径的代理
|
||
- **[nextlevelbuilder/ui-ux-pro-max-skill](https://github.com/nextlevelbuilder/ui-ux-pro-max-skill)** - UI/UX 设计模式与最佳实践
|
||
- **[ehmo/platform-design-skills](https://github.com/ehmo/platform-design-skills)** - 面向跨平台应用的 300+ 条设计规则,源自 Apple HIG、Material Design 3 与 WCAG 2.2
|
||
- **[scarletkc/vexor](https://github.com/scarletkc/vexor)** - 基于向量的 CLI,用于语义文件搜索,并提供 Claude/Codex skill
|
||
- **[obra/test-driven-development](https://github.com/obra/superpowers/blob/main/skills/test-driven-development/SKILL.md)** - 在实现代码之前编写测试
|
||
- **[obra/subagent-driven-development](https://github.com/obra/superpowers/blob/main/skills/subagent-driven-development/SKILL.md)** - 使用多个子代理进行开发
|
||
- **[obra/systematic-debugging](https://github.com/obra/superpowers/blob/main/skills/systematic-debugging/SKILL.md)** - 在代码中进行系统化的问题排查
|
||
- **[obra/root-cause-tracing](https://github.com/obra/superpowers/blob/main/skills/root-cause-tracing/SKILL.md)** - 调查并识别根本问题
|
||
- **[obra/testing-skills-with-subagents](https://github.com/obra/superpowers/blob/main/skills/testing-skills-with-subagents/SKILL.md)** - 协作式测试方法
|
||
- **[obra/testing-anti-patterns](https://github.com/obra/superpowers/blob/main/skills/testing-anti-patterns/SKILL.md)** - 识别无效的测试实践
|
||
- **[obra/finishing-a-development-branch](https://github.com/obra/superpowers/blob/main/skills/finishing-a-development-branch/SKILL.md)** - 完成 Git 代码分支
|
||
- **[obra/requesting-code-review](https://github.com/obra/superpowers/blob/main/skills/requesting-code-review/SKILL.md)** - 发起代码审查流程
|
||
- **[obra/receiving-code-review](https://github.com/obra/superpowers/blob/main/skills/receiving-code-review/SKILL.md)** - 处理并吸收代码反馈
|
||
- **[obra/using-git-worktrees](https://github.com/obra/superpowers/blob/main/skills/using-git-worktrees/SKILL.md)** - 管理多个 Git 工作树
|
||
- **[obra/verification-before-completion](https://github.com/obra/superpowers/blob/main/skills/verification-before-completion/SKILL.md)** - 在最终定稿前验证工作成果
|
||
- **[obra/condition-based-waiting](https://github.com/obra/superpowers/blob/main/skills/condition-based-waiting/SKILL.md)** - 管理基于条件的暂停或延迟
|
||
- **[obra/commands](https://github.com/obra/superpowers/tree/main/skills/commands)** - 创建并管理命令结构
|
||
- **[obra/writing-skills](https://github.com/obra/superpowers/blob/main/skills/writing-skills/SKILL.md)** - 开发并记录能力
|
||
- **[fvadicamo/dev-agent-skills](https://github.com/fvadicamo/dev-agent-skills)** - 面向提交、PR 与代码审查的 Git 与 GitHub 工作流技能
|
||
- **[omkamal/pypict-skill](https://github.com/omkamal/pypict-claude-skill/blob/main/SKILL.md)** - 成对测试生成
|
||
- **[alinaqi/claude-bootstrap](https://github.com/alinaqi/claude-bootstrap)** - 带有明确主张的项目初始化,包含安全优先护栏、规格驱动的原子化待办、LLM 测试模式,以及 CLI 工具编排(gh、vercel、supabase)
|
||
- **[ZhangHanDong/makepad-skills](https://github.com/ZhangHanDong/makepad-skills)** - 面向 Rust 应用的 Makepad UI 开发技能:搭建、模式、着色器、打包与故障排查。
|
||
- **[massimodeluisa/recursive-decomposition-skill](https://github.com/massimodeluisa/recursive-decomposition-skill)** - 通过基于 RLM 研究的递归分解策略处理长上下文任务(100+ 文件、50k+ token)
|
||
- **[AvdLee/swiftui-expert-skill](https://github.com/AvdLee/SwiftUI-Agent-Skill/tree/main/swiftui-expert-skill)** - 现代 SwiftUI 最佳实践与 iOS 26+ Liquid Glass 适配
|
||
- **[efremidze/swift-patterns-skill](https://github.com/efremidze/swift-patterns-skill/tree/main/swift-patterns)** - 现代 Swift/SwiftUI 最佳实践
|
||
- **[Joannis/claude-skills](https://github.com/Joannis/claude-skills)** - Swift Server 开发指南,附带用于最佳实践的 lint 工具
|
||
- **[rudrankriyam/app-store-connect-cli-skills](https://github.com/rudrankriyam/app-store-connect-cli-skills)** - 使用 ASC CLI 自动化 App Store 部署与管理
|
||
- **[rameerez/claude-code-startup-skills](https://github.com/rameerez/claude-code-startup-skills)** - 用于构建并运营软件初创公司、应用与 SaaS 的技能
|
||
- **[zscole/model-hierarchy-skill](https://github.com/zscole/model-hierarchy-skill)** - 基于任务复杂度的成本优化模型路由
|
||
- **[CloudAI-X/threejs-skills](https://github.com/CloudAI-X/threejs-skills)** - 用于创建 3D 元素与交互体验的 Three.js 技能
|
||
- **[Leonxlnx/taste-skill](https://github.com/Leonxlnx/taste-skill)** - 高自主性前端技能,通过可调的设计差异、动效强度与视觉密度赋予 AI 良好品味,避免千篇一律的 UI
|
||
- **[testdino-hq/playwright-skill](https://github.com/testdino-hq/playwright-skill)** - 70+ 条经生产验证的 Playwright 自动化测试模式:E2E、POM、CI/CD、迁移、CLI
|
||
- **[NeoLabHQ/code-review](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/code-review)** - 使用专用代理进行全面 PR 代码审查:bug-hunter、security-auditor、code-quality-reviewer、contracts-reviewer、historical-context-reviewer、test-coverage-reviewer
|
||
- **[NeoLabHQ/reflexion](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/reflexion)** - 自我精炼循环,迫使 LLM 反思先前输出并自我纠正。
|
||
- **[NeoLabHQ/sdd](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/sdd)** - 规格驱动开发工作流,通过结构化规划、架构设计与基于 LLM-as-a-Judge 的质量门禁,将提示转化为可上线的实现。
|
||
- **[NeoLabHQ/ddd](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/ddd)** - 领域驱动开发技能,同时涵盖 Clean Architecture、SOLID 原则与设计模式。
|
||
- **[NeoLabHQ/sadd](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/sadd)** - 为各项任务分派独立子代理,在迭代之间设置代码审查检查点,以实现快速且可控的开发。
|
||
- **[NeoLabHQ/kaizen](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/kaizen)** - 基于日式 Kaizen 哲学与精益方法论,应用持续改进方法论与多种分析方法。
|
||
- **[hamelsmu/eval-audit](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/eval-audit)** - 审计 LLM 评估流水线并暴露问题
|
||
- **[hamelsmu/error-analysis](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/error-analysis)** - 系统化识别 LLM 流水线中的失败模式
|
||
- **[hamelsmu/generate-synthetic-data](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/generate-synthetic-data)** - 为 LLM 评估创建多样化的合成测试输入
|
||
- **[hamelsmu/write-judge-prompt](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/write-judge-prompt)** - 为主观标准设计 LLM-as-a-Judge 评估器
|
||
- **[hamelsmu/validate-evaluator](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/validate-evaluator)** - 根据人工标注校准 LLM 评判器
|
||
- **[hamelsmu/evaluate-rag](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/evaluate-rag)** - 评估 RAG 检索与生成质量
|
||
- **[uucz/moyu](https://github.com/uucz/moyu)** - 反过度工程化技能,含 5 种变体与 10 个平台
|
||
- **[hamelsmu/build-review-interface](https://github.com/hamelsmu/prompts/tree/main/evals-skills/skills/build-review-interface)** - 构建用于审查 LLM 追踪记录的标注界面
|
||
- **[mattpocock/skills](https://github.com/mattpocock/skills)** - 17 项开发工作流技能:PRD 撰写、TDD、代码库架构、Git 护栏、Issue 分诊、重构计划等
|
||
- **[mukul975/Anthropic-Cybersecurity-Skills](https://github.com/mukul975/Anthropic-Cybersecurity-Skills)** - 跨 38 个领域的 753 项网络安全技能:云安全、渗透测试、红队、DFIR、恶意软件分析、威胁情报等(映射 MITRE ATT&CK)
|
||
- **[wrsmith108/varlock-claude-skill](https://github.com/wrsmith108/varlock-claude-skill)** - 安全的环境变量管理,确保密钥绝不会暴露在 Claude 会话、终端、日志或 git 提交中
|
||
- **[Skill_Seekers](https://github.com/yusufkaraaslan/Skill_Seekers)** - 在数分钟内自动将文档网站、GitHub 仓库与 PDF 转换为 Claude AI 技能
|
||
- **[NoizAI/skills](https://github.com/NoizAI/skills)** - 拟人化 TTS 工作流,支持本地/云端 API 与应用交付
|
||
- **[Kevin7Qi/codex-collab](https://github.com/Kevin7Qi/codex-collab)** - 从 Claude Code 与 Codex 协作
|
||
- **[ethos-link/rails-conventions](https://github.com/ethos-link/rails-conventions)** - Rails 8 约定,用于保持一致的生产代码变更
|
||
- **[ShunsukeHayashi/agent-skill-bus](https://github.com/ShunsukeHayashi/agent-skill-bus)** - 面向 AI 代理系统的自我改进型任务编排
|
||
- **[mcollina/skills](https://github.com/mcollina/skills/tree/main/skills)** - Matteo Collina 的 11 项技能:Node.js、Fastify、TypeScript、OAuth、Git/GitHub、ESLint neostandard、文档(Diataxis)、Node.js 核心内部机制、skill 优化器等
|
||
- **[Lum1104/understand-anything](https://github.com/Lum1104/Understand-Anything)** - 通过多代理 LLM 分析构建交互式代码库知识图谱
|
||
- **[hqhq1025/skill-optimizer](https://github.com/hqhq1025/skill-optimizer)** - 基于真实会话数据与研究支撑的静态分析,诊断并优化 Agent Skills(SKILL.md)。兼容 Claude Code、Codex 及任何兼容 Agent Skills 的代理
|
||
- **[LambdaTest/agent-skills](https://github.com/LambdaTest/agent-skills)** - TestMu AI(前身为 LambdaTest)Skills 是精选的 Agent Skills 集合,教导 AI 编码助手如何编写生产级测试自动化。
|
||
- **[foryourhealth111-pixel/Vibe-Skills](https://github.com/foryourhealth111-pixel/Vibe-Skills)** - 受技能治理的即插即用框架,用于分阶段、测试驱动的技能编排
|
||
- **[metalbear-co/skills](https://github.com/metalbear-co/skills)** - 让代理使用 mirrord 针对你的 Kubernetes 集群编写代码并测试的技能
|
||
- **[dembrandt/dembrandt-skills](https://github.com/dembrandt/dembrandt-skills)** - UX 与设计系统技能:层级、字体排版、无障碍、交互
|
||
- **[GanyuanRan/Aegis](https://github.com/GanyuanRan/Aegis)** - 面向 AI 编码代理的证据驱动方法包
|
||
- **[baskduf/codex-fable5](https://github.com/baskduf/FableCodex/tree/main/plugins/codex-fable5/skills/codex-fable5)** - 面向 Codex 的基于证据的工作流门禁
|
||
- **[csthink/dashmotion](https://github.com/csthink/dashmotion/tree/main/skills/dashmotion)** - 从纯英文或 Mermaid 生成动画技术图表,输出自包含 HTML/SVG
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">上下文工程(Context Engineering)</h3></summary>
|
||
|
||
- **[muratcankoylan/context-fundamentals](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/context-fundamentals)** - 理解什么是上下文、为何重要,以及智能体(agent)系统中上下文的构成
|
||
- **[muratcankoylan/context-degradation](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/context-degradation)** - 识别上下文失效模式:lost-in-middle(中间迷失)、poisoning(污染)、distraction(干扰)与 clash(冲突)
|
||
- **[muratcankoylan/context-compression](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/context-compression)** - 为长期运行的会话设计并评估压缩策略
|
||
- **[muratcankoylan/context-optimization](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/context-optimization)** - 应用压缩(compaction)、掩码(masking)与缓存策略
|
||
- **[muratcankoylan/multi-agent-patterns](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/multi-agent-patterns)** - 掌握编排器(orchestrator)、点对点(peer-to-peer)与层级式多智能体架构
|
||
- **[muratcankoylan/memory-systems](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/memory-systems)** - 设计短期、长期与基于图(graph-based)的记忆架构
|
||
- **[muratcankoylan/tool-design](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/tool-design)** - 构建智能体能高效使用的工具,包括架构简化(architectural reduction)模式
|
||
- **[muratcankoylan/evaluation](https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering/tree/main/skills/evaluation)** - 为智能体系统构建评估框架
|
||
- **[k-kolomeitsev/data-structure-protocol](https://github.com/k-kolomeitsev/data-structure-protocol)** - 面向 AI(LLM)编码智能体的基于图的长期记忆技能 — 更快的上下文、更少的 token、更安全的重构
|
||
- **[awrshift/claude-memory-kit](https://github.com/awrshift/claude-memory-kit)** - 面向多项目工作流的持久化记忆:hooks、wiki 与每日综合汇总
|
||
- **[NeoLabHQ/prompt-engineering](https://github.com/NeoLabHQ/context-engineering-kit/tree/master/plugins/customaize-agent/skills/prompt-engineering)** - 广泛使用的提示工程(prompt engineering)技术与模式,包括 Anthropic 最佳实践与智能体说服(agent persuasion)原则。
|
||
- **[sametbrr/llm-wiki-manager](https://github.com/sametbrr/llm-wiki-manager)** - 由 LLM 管理的持久化个人 wiki — 你负责策展来源,模型负责撰写、交叉引用与维护知识库。以 8 种运行模式实现 Karpathy 的 LLM Wiki 模式。
|
||
- **[Panniantong/Agent-Reach](https://github.com/Panniantong/Agent-Reach)** - 覆盖 17 个站点(含中文平台)的多平台搜索 CLI
|
||
- **[Tubo2333/obsidian-knowledge-brain](https://github.com/Tubo2333/obsidian-knowledge-brain)** - 面向 AI 编码智能体的跨会话知识记忆与规则演化
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">专业领域</h3></summary>
|
||
|
||
- **[shouldnotappearcalm/a-share-skill](https://github.com/shouldnotappearcalm/a-share-skill)** - 中国 A 股(沪/深)技能:实时行情、K 线历史、技术指标、事件、资金流向、板块热力图与模拟交易。兼容 Claude Code、Cursor、Codex 与 Qoder
|
||
- **[transloadit/skills](https://github.com/transloadit/skills/tree/main/skills)** - Transloadit 技能合集(6 项)
|
||
- **[honeydew-ai/honeydew-ai-coding-agents-plugins](https://github.com/honeydew-ai/honeydew-ai-coding-agents-plugins)** - 面向 Snowflake、Databricks 与 BigQuery 上 Honeydew 语义层的 11 项技能:模型探索,实体/关系/属性/指标/上下文/领域创建,验证、查询、筛选与工作区分支
|
||
- **[raintree-technology/apple-hig-skills](https://github.com/raintree-technology/apple-hig-skills)** - 将 Apple Human Interface Guidelines 拆分为 14 项智能体技能,涵盖 iOS、macOS、visionOS、watchOS 与 tvOS 的平台、基础、组件、模式、输入与技术
|
||
- **[K-Dense-AI/claude-scientific-skills](https://github.com/K-Dense-AI/claude-scientific-skills)** - 科学研究与分析技能
|
||
- **[NotMyself/claude-win11-speckit-update-skill](https://github.com/NotMyself/claude-win11-speckit-update-skill)** - Windows 11 系统管理
|
||
- **[sanjay3290/imagen](https://github.com/sanjay3290/ai-skills/tree/main/skills/imagen)** - 使用 Google Gemini API 生成图像
|
||
- **[jeffersonwarrior/claudisms](https://github.com/jeffersonwarrior/claudisms)** - 短信(SMS)消息集成
|
||
- **[SHADOWPR0/security-bluebook-builder](https://github.com/SHADOWPR0/security-bluebook-builder)** - 为敏感应用构建安全 Blue Book
|
||
- **[obra/defense-in-depth](https://github.com/obra/superpowers/blob/main/skills/defense-in-depth/SKILL.md)** - 多层纵深安全(defense in depth)方案
|
||
- **[huifer/Claude-Ally-Health](https://github.com/huifer/Claude-Ally-Health)** - 健康助手技能:医疗信息分析、症状追踪与健康指导。
|
||
- **[frmoretto/clarity-gate](https://github.com/frmoretto/clarity-gate)** - 面向 RAG 系统的认知质量(epistemic quality)验证
|
||
- **[wanshuiyin/Auto-claude-code-research-in-sleep](https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep)** - 自主机器学习研究:跨模型评审循环与 GPU 部署
|
||
- **[zechenzhangAGI/AI-research-SKILLs](https://github.com/zechenzhangAGI/AI-research-SKILLs)** - 77 项 AI 研究技能,覆盖模型训练、推理与 MLOps
|
||
- **[Orchestra-Research/AI-research-SKILLs](https://github.com/Orchestra-Research/AI-research-SKILLs)** - 20 模块 AI 研究技能库:模型架构、训练与 ML 论文写作
|
||
- **[komal-SkyNET/claude-skill-homeassistant](https://github.com/komal-SkyNET/claude-skill-homeassistant)** - 增强并管理 Home Assistant 工作流
|
||
- **[more-io/apple-bridges](https://github.com/more-io/claude-apple-bridges)** - 原生 macOS 应用访问 — 通过 Swift CLI 桥接管理 Apple 提醒事项、日历、通讯录、备忘录、邮件与 tmux 会话
|
||
- **[hanhuark/mechanical-engineering-research-skill](https://github.com/hanhuark/mechanical-engineering-research-skill)** - 热流体研究写作、申请书、DOE 与答辩反馈
|
||
- **[prompt-security/clawsec](https://github.com/prompt-security/clawsec)** - 安全技能套件:漂移检测、自动化审计与技能完整性校验
|
||
- **[BehiSecc/vibesec](https://github.com/BehiSecc/VibeSec-Skill)** - 以漏洞猎手视角帮助编写安全代码,防范 IDOR、XSS、SQL 注入、SSRF 与弱认证等常见漏洞
|
||
- **[lawvable/awesome-legal-skills](https://github.com/lawvable/awesome-legal-skills)** - 精选智能体技能,用于自动化法律工作流
|
||
- **[peas/genealogy-research](https://paulo.com.br/skills/genealogy-research/SKILL.md)** - 族谱研究智能体:OCR、FamilySearch、YAML 数据与人机协同(human-in-the-loop)
|
||
- **[zw008/VMware-AIops](https://github.com/zw008/VMware-AIops)** - AI 驱动的 VMware vCenter/ESXi 监控与运维:清单查询、健康/告警、虚拟机生命周期(创建、删除、快照、克隆、迁移)、vSAN 管理、Aria Operations 分析与定时日志扫描。支持 Claude Code、Gemini CLI、Codex、Aider、Trae、Kimi 与 MCP。
|
||
- **[video-db/skills](https://github.com/video-db/skills)** - 实时与批量视频工作流:采集屏幕/音频,摄取 URL/YouTube/RTSP,转写、索引、搜索,生成字幕,编辑时间线,并流式输出 HLS
|
||
- **[materials-simulation-skills](https://github.com/HeshamFS/materials-simulation-skills)** - 计算材料科学智能体技能:数值稳定性、时间步进、线性求解器、网格生成、仿真验证、参数优化与后处理
|
||
- **[Ericyoung-183/alpha-insights](https://github.com/Ericyoung-183/alpha-insights)** - 面向 Claude Code 与 Codex 的 Harness 强制业务研究
|
||
- **[takechanman1228/claude-ecom](https://github.com/takechanman1228/claude-ecom)** - 电商 CSV 转业务复盘,含 KPI 分解
|
||
- **[talkstream/ru-text](https://github.com/talkstream/ru-text)** - 俄语文本质量:约 1,040 条排版、信息风格、编辑、UX 写作与商务信函规则。跨平台:Claude Code、Codex CLI、Gemini CLI、Cursor。
|
||
- **[helius-labs/helius-skills](https://github.com/helius-labs/core-ai/tree/main/helius-skills)** - 端到端交付 Solana 应用;交易发送、资产查询、实时流、代币兑换、预测市场、浏览器钱包,以及基于 Helius API、DFlow 交易与 Phantom 钱包集成的协议内部深度研究
|
||
- **[meodai/skill.color-expert](https://github.com/meodai/skill.color-expert)** - 色彩科学专家技能,含 286K 词参考资料,覆盖 OKLCH/OKLAB、调色板生成、可访问性/对比度、色彩命名、颜料混合与历史色彩理论
|
||
- **[aklofas/kicad-happy](https://github.com/aklofas/kicad-happy)** - AI 驱动的 KiCad 电子设计评审与分析
|
||
- **[bitwize-music-studio/claude-ai-music-skills](https://github.com/bitwize-music-studio/claude-ai-music-skills)** - AI 音乐专辑全生命周期制作
|
||
- **[Alisa0808/vibe-creating-skill](https://github.com/Alisa0808/vibe-creating-skill)** - 将粗略创意或分镜脚本改写为文生视频(text-to-video)提示词
|
||
|
||
</details>
|
||
|
||
</details>
|
||
|
||
<details>
|
||
<summary><h3 style="display:inline">n8n Automation</h3></summary>
|
||
|
||
- **[czlonkowski/n8n-code-javascript](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-code-javascript)** - 在 n8n Code 节点中使用 JavaScript,涵盖数据访问模式
|
||
- **[czlonkowski/n8n-code-python](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-code-python)** - 在 n8n Code 节点中使用 Python,并说明相关限制
|
||
- **[czlonkowski/n8n-expression-syntax](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-expression-syntax)** - n8n 表达式语法,包括 {{}} 以及 $json/$node 变量
|
||
- **[czlonkowski/n8n-mcp-tools-expert](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-mcp-tools-expert)** - MCP 工具指南,涵盖工具选择与节点格式
|
||
- **[czlonkowski/n8n-node-configuration](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-node-configuration)** - 节点配置,包括依赖规则与 AI 连接
|
||
- **[czlonkowski/n8n-validation-expert](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-validation-expert)** - 修复 n8n 校验错误,附错误目录
|
||
- **[czlonkowski/n8n-workflow-patterns](https://github.com/czlonkowski/n8n-skills/tree/main/skills/n8n-workflow-patterns)** - 工作流模式,适用于 webhook、HTTP、数据库与 AI 任务
|
||
|
||
</details>
|
||
|
||
|
||
|
||
|
||
## 🔒 安全提示
|
||
|
||
本列表中的 Skills 经过精选,但未经过安全审计。在被收录后,原作者维护者可能随时对其进行更新、修改或替换。
|
||
|
||
在安装或使用任何 Agent Skill 之前,请自行评估潜在安全风险并验证来源。
|
||
|
||
推荐工具:
|
||
|
||
- [Synk Skill Security Scanner](https://github.com/snyk/agent-scan)
|
||
- [Agent Trust Hub](https://ai.gendigital.com/agent-trust-hub)
|
||
|
||
Agent Skill 可能包含提示注入(prompt injection)、工具投毒(tool poisoning)、隐藏的恶意软件载荷,或不安全的数据处理模式。请务必审查代码,并自行斟酌后使用。
|
||
|
||
|
||
## 其他 AI 编程助手的 Skills 路径
|
||
|
||
| Tool | Project Path | Global Path | Official Docs |
|
||
|------|-------------|-------------|---------------|
|
||
| Antigravity | `.agent/skills/` | `~/.gemini/antigravity/skills/` | [Antigravity Skills](https://antigravity.google/docs/skills) |
|
||
| Claude Code | `.claude/skills/` | `~/.claude/skills/` | [Claude Code Skills](https://docs.anthropic.com/en/docs/claude-code/skills) |
|
||
| Codex | `.agents/skills/` | `~/.agents/skills/` | [Codex Skills](https://developers.openai.com/codex/skills) |
|
||
| Cursor | `.cursor/skills/` | `~/.cursor/skills/` | [Cursor Skills](https://cursor.com/docs/context/skills) |
|
||
| Gemini CLI | `.gemini/skills/` | `~/.gemini/skills/` | [Gemini CLI Skills](https://geminicli.com/docs/cli/skills/) |
|
||
| GitHub Copilot | `.github/skills/` | `~/.copilot/skills/` | [Copilot Skills](https://docs.github.com/en/copilot/concepts/agents/about-agent-skills) |
|
||
| OpenCode | `.opencode/skills/` | `~/.config/opencode/skills/` | [OpenCode Skills](https://opencode.ai/docs/skills) |
|
||
| Windsurf | `.windsurf/skills/` | `~/.codeium/windsurf/skills/` | [Windsurf Cascade Skills](https://docs.windsurf.com/windsurf/cascade/skills) |
|
||
|
||
|
||
## Skill 质量标准
|
||
|
||
随着生态不断壮大,一致的质量标准有助于 Agent 可靠地发现并使用 Skills。以下参考资料与准则有助于维持较高水准。
|
||
|
||
|
||
### 质量标准
|
||
|
||
| Area | Guideline |
|
||
|------|-----------|
|
||
| **Description** | 使用第三人称撰写。说明该 Skill *做什么*以及*何时*使用。使用 Agent 可匹配的具体关键词(例如用 "PostgreSQL migration" 而非 "database stuff")。 |
|
||
| **Progressive disclosure** | 顶层元数据控制在约 100 tokens 以内。Skill 正文应少于 500 行。按需加载资源(大型文档、schema),不要内联嵌入。 |
|
||
| **No absolute paths** | 切勿硬编码机器相关路径,例如 `/Users/alice/`。应使用相对路径或公认变量(`$HOME`、`$PROJECT_ROOT`)。 |
|
||
| **Scoped tools** | 仅申请该 Skill 实际需要的工具。避免笼统使用 `"tools": ["*"]`。应明确声明工具依赖。 |
|
||
|
||
|
||
## 🤝 贡献
|
||
|
||
欢迎贡献!请参阅 [CONTRIBUTING.md](CONTRIBUTING.md) 了解指南。
|
||
|
||
- 通过 PR 提交新 Skills
|
||
- 改进现有定义
|
||
|
||
**注意:** 请勿提交你 3 小时前才创建的 Skills。我们目前更关注社区已采纳的 Skills,尤其是由开发团队发布并在实际使用中验证过的项目。质量优先于数量。
|
||
|
||
## 贡献者 ♥️ 致谢
|
||

|
||
|
||
## 许可证
|
||
|
||
MIT License - 详见 [LICENSE](LICENSE)
|
||
|
||
这是一份精选列表。此处列出的 Skills 由各自作者与团队创建和维护,并非由我们创建。我们筛选的是社区已采纳、经过验证的 Skills,不对所列项目进行审计、背书或就其安全性与正确性作出保证。它们未经安全审计,在生产环境使用前应自行审查。
|
||
|
||
若你发现某 listed Skill 存在问题,或希望移除你的 Skill,请 [open an issue](https://github.com/VoltAgent/awesome-agent-skills/issues),我们会尽快处理。
|
||
|
||
[codex-badge]: https://img.shields.io/github/stars/VoltAgent/awesome-codex-subagents?style=classic&label=Codex%20Subagents&color=000000&logo=data:image/svg%2bxml;base64,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
|
||
[codex-link]: https://github.com/VoltAgent/awesome-codex-subagents
|