安全与隐私 | Agent Skills
安全与隐私
安全与隐私类技能用于辅助识别风险、检查访问控制与合规要求,并按安全领域规则完成工作。
目录快照:2026 年 11 月 2 日
分类目录
- addyosmani/accessibility
- addyosmani/best-practices
- alinaqi/maggy
- aurite-ai/agent-verifier
- auth0/auth0-android
- auth0/auth0-angular
- auth0/auth0-express
- auth0/auth0-fastify
- auth0/auth0-mfa
- auth0/auth0-migration
- auth0/auth0-nextjs
- auth0/auth0-nuxt
- auth0/auth0-react
- auth0/auth0-react-native
- auth0/auth0-vue
- BehiSecc/vibesec
- better-auth/best-practices
- better-auth/create-auth
- better-auth/emailAndPassword
- better-auth/explain-error
- better-auth/organization
- better-auth/providers
- better-auth/twoFactor
- binance/query-token-audit
- coinbase/authenticate-wallet
- composiohq/composio
- danoszz/penthera
- firebase/firebase-auth-basics
- firebase/firebase-basics
- firebase/firebase-security-rules-auditor
- firecrawl/firecrawl-build-interact
- garrytan/cso
- google/cloud/google-cloud-recipe-auth
- google/cloud/google-cloud-waf-security
- mcollina/skills
- microsoft/azure-identity-dotnet
- microsoft/azure-identity-java
- microsoft/azure-identity-py
- microsoft/azure-identity-rust
- microsoft/azure-identity-ts
- microsoft/azure-keyvault-py
- microsoft/azure-keyvault-secrets-rust
- microsoft/azure-keyvault-secrets-ts
- microsoft/azure-security-keyvault-keys-dotnet
- microsoft/azure-security-keyvault-keys-java
- microsoft/azure-security-keyvault-secrets-java
- microsoft/fastapi-router-py
- microsoft/microsoft-azure-webjobs-extensions-authentication-events-dotnet
- MiniMax-AI/fullstack-dev
- MiniMax-AI/ios-application-dev
- mongodb/mongodb-mcp-setup
- moxno/privacyscrubber-guard
- mukul975/Anthropic-Cybersecurity-Skills
- NeoLabHQ/review
- openai/netlify-deploy
- openai/security-best-practices
- openai/security-ownership-map
- openai/security-threat-model
- prompt-security/clawsec
- redhat/cve-skillpack
- Rootly-AI-Labs/rootly-incident-responder
- saleh-alhaddad/itqan-engineering
- SHADOWPR0/security-bluebook-builder
- squirrelscan/squirrelscan
- Tencent/aig-agent-redteam
- trailofbits/ask-questions-if-underspecified
- trailofbits/audit-context-building
- trailofbits/building-secure-contracts
- trailofbits/burpsuite-project-parser
- trailofbits/claude-in-chrome-troubleshooting
- trailofbits/constant-time-analysis
- trailofbits/differential-review
- trailofbits/dwarf-expert
- trailofbits/entry-point-analyzer
- trailofbits/firebase-apk-scanner
- trailofbits/insecure-defaults
- trailofbits/modern-python
- trailofbits/property-based-testing
- trailofbits/semgrep-rule-creator
- trailofbits/semgrep-rule-variant-creator
- trailofbits/sharp-edges
- trailofbits/spec-to-code-compliance
- trailofbits/static-analysis
- trailofbits/testing-handbook-skills
- trailofbits/variant-analysis
- veniceai/venice-api-overview
- veniceai/venice-auth
- vercel-labs/vercel-cli-with-tokens
- VoDaiLocz/kilo-kit-mcp
- WordPress/wp-abilities-api
- WordPress/wp-plugin-development
- WordPress/wp-rest-api
- wrsmith108/varlock-claude-skill
本类全部技能
addyosmani/accessibility
目录来源: Skills by - Google Chrome team - Addy Osmani (Web Quality)
来源文档: https://github.com/addyosmani/web-quality-skills/blob/main/skills/accessibility/SKILL.md
功能简介: WCAG合规性、屏幕阅读器支持和键盘导航
适用时机: 在被要求“提高可访问性”、“a11y审计”、“WCAG合规性”、“屏幕阅读器支持”、“键盘导航”或“使可访问”时使用。
来源技能章节菜单:
- Accessibility (a11y)
- Evidence-led audit workflow
- WCAG Principles: POUR
- Conformance levels
- Perceivable
- Text alternatives (1.1)
- Color contrast (1.4.3, 1.4.6)
- Media alternatives (1.2)
- Operable
- Keyboard accessible (2.1)
- Focus visible (2.4.7)
- Focus not obscured (2.4.11) — new in 2.2
- Skip links (2.4.1)
- Target size (2.5.8) — new in 2.2
- Dragging movements (2.5.7) — new in 2.2
- Timing (2.2)
addyosmani/best-practices
目录来源: Skills by - Google Chrome team - Addy Osmani (Web Quality)
来源文档: https://github.com/addyosmani/web-quality-skills/blob/main/skills/best-practices/SKILL.md
功能简介: 安全、现代网络API和代码质量模式
适用时机: 在被要求“应用最佳实践”、“安全审计”、“现代化代码”、“代码质量审查”或“检查漏洞”时使用。
来源技能章节菜单:
- Best practices
- Evidence-led audit workflow
- Security
- Browser compatibility
- Doctype declaration
- Character encoding
- Viewport meta tag
- Feature detection
- Polyfills (when needed)
- Deprecated APIs
- Avoid these
- Event listener passive
- Console & errors
- No console errors
- Error boundaries (React)
- Global error handler
alinaqi/maggy
目录来源: Development and Testing
来源文档: https://github.com/alinaqi/maggy/blob/main/skills/maggy/SKILL.md
功能简介: 采用安全优先的防护框架、以规格驱动的原子待办事项、大型语言模型测试模式和命令行工具编排(gh、vercel、supabase)的个性化项目初始化。
适用时机: 当任务需要此目录条目中描述的能力时使用:采用安全优先的防护措施、以规范为驱动的原子待办事项、大型语言模型测试模式和命令行工具编排(gh、vercel、supabase)的强硬项目初始化。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Maggy Skill
- ⚠️ Execute permission model (important)
- When Maggy Helps
- Install and Configure
- Provider Abstraction
- Execute Pipeline
- Competitor Intelligence
- Not Included
- Files
aurite-ai/agent-verifier
目录来源: Development and Testing
来源文档: https://github.com/Aurite-ai/agent-verifier/blob/main/README.md
功能简介: 审查AI代理代码中的秘密、无限循环和幻觉工具
适用时机: 当任务需要此目录条目中描述的能力时使用:审查AI代理代码中的秘密、无限循环、幻觉工具
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- 🔍 Aurite Agent Verifier
- Why?
- Author’s Note
- What You Get
- Quickstart
- Install as a Claude Code plugin
- Learn More
- Contributions
- Table of Contents
- Architecture
- Available Skills
- Installation
- Recommended
- Updating Installed Skills
- From NPM Registry
- Usage
此目录条目指向一个技能集合。扫描来源仓库发现 5 个不重复的技能文件路径。
部分成员: verification · verify-language · verify-patterns · verify-quality · verify-security
auth0/auth0-android
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 使用Auth0 SDK为原生Android应用程序添加身份验证
适用时机: 在带有 Auth0 Web Auth 的 Android 应用中添加登录和退出屏幕 在 Android 上使用生物识别保护安全存储用户令牌 在 Kotlin 移动应用中设置无密码认证 在 Android 项目中处理多因素认证流程 使用 Auth0 访问令牌将 Android 应用连接到受保护的 API
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-angular
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 使用@auth0/auth0-angular为Angular应用程序添加身份验证
适用时机: 在基于重定向的身份验证的 Angular SPA 中添加登录和注册功能 保护特定路由,确保仅限已登录的用户访问 自动将访问令牌附加到出站 API 调用中 登录后自动显示用户个人资料数据(如姓名和头像) 使用 Angular 可观察性对象(observables)根据身份验证状态控制功能访问
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-express
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为Express.js应用程序添加基于会话的身份验证
适用时机: 在不手动编写 OAuth 流程的情况下,为 Express.js 网页应用添加登录和退出功能 保护管理员或仪表板路由,确保仅限已认证用户访问 路由处理器中从 Auth0 会话中读取用户个人资料数据 为 Express 设置 OpenID Connect 流程的回调和重定向 URL 限制访问特定 Express 路由,同时保持主页公开
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-fastify
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为Fastify网络应用程序添加基于会话的身份验证
适用时机: 在不手动编写 OAuth 流程的情况下,为 Fastify 网页应用添加登录和退出功能 保护特定路由,确保仅限已认证用户访问 检索已登录用户的个人资料以显示在仪表盘上 从会话中获取访问令牌以调用下游 API 在 Fastify 中为 Auth0 的托管登录页面设置回调处理
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-mfa
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为由Auth0驱动的应用添加多因素认证
适用时机: 将TOTP或短信验证添加到现有的Auth0登录流程 在用户更改付款信息之前要求实施增强型多因素认证(MFA) 设置适应性认证,仅在可疑登录时触发MFA 通过在每次会话中强制实施多因素认证来满足PCI-DSS或HIPAA合规要求 将WebAuthn凭据配置为无密码应用的可用第二因素
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-migration
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 将用户和身份验证流程从其他提供商迁移到Auth0
适用时机: 将 Firebase Auth 用户基础库迁移到 Auth0,但不强制重置密码 在分阶段迁移期间并行运行身份验证 从自定义数据库批量导入用户到 Auth0,更新 Express.js API 的 JWT 验证从 HS256 更改为 Auth0 的 RS256 在用户下次登录时懒惰迁移活跃的 Supabase 用户到 Auth0
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-nextjs
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为Next.js应用程序添加身份验证
适用时机: 在 Next.js App Router 项目中添加登录和退出流程 保护特定页面,确保仅限已认证用户访问 在服务器组件和 API 路由中读取用户会话数据 使用 Auth0 中件设置 OAuth 回调处理 将 auth 路由从 SDK v3 迁移到 v4 格式
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-nuxt
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 在使用加密cookie会话的Nuxt 3/4应用程序中添加Auth0认证
适用时机: 在带有 Auth0 的 Nuxt 3 应用中添加登录和退出流程 通过会话检查保护服务器路由和 API 端点 获取访问令牌以代表用户调用外部 API 使用 Redis 设置状态会话以实现分布式 Nuxt 部署 在 Nuxt SSR 应用中配置基于角色的路由中间件
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-react
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 使用@auth0/auth0-react为React SPA添加身份验证
适用时机: 在 React SPA 中添加登录和退出按钮 保护特定路由,确保仅限已登录用户访问 调用带有 Auth0 访问令牌的后端 API 在 React 应用中处理多因素身份验证(MFA)注册和验证 为 Vite 或 CRA 设置 Auth0Provider 的正确环境变量
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-react-native
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为React Native和Expo移动应用程序添加身份验证
适用时机: 在带有 Auth0 通用登录功能的 React Native 应用中添加登录和注册界面 在 iOS 和 Android 上为返回用户设置生物识别认证 从移动应用调用受保护的 API,使用已存储的访问令牌 在 Expo 项目中配置 OAuth 重定向的深度链接回调 将现有 React Native 应用从自定义认证迁移到 Auth0
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
auth0/auth0-vue
目录来源: Skills by Auth0
来源文档: https://github.com/auth0/agent-skills/blob/main/README.md
功能简介: 为Vue.js应用程序添加身份验证
适用时机: 在 Vue 3 SPA 中添加登录和退出按钮 使用需要身份验证的导航守卫保护特定路由 代表已登录用户调用后端 API 获取访问令牌 登录后显示用户资料信息,如姓名和头像 为新 Vue 项目设置 Auth0 重定向 URI 和环境变量
来源技能章节菜单:
- Auth0 Agent Skills
- Prerequisites
- Install
- Claude Code
- Cursor
- Codex / ChatGPT plugins
- Antigravity
- Gemini CLI
- Any Agent (Skills CLI)
- ClawHub Marketplace
- What happens after install
- Forcing the skill with `/auth0`
- Migrating from the individual skills
- Coverage
- Example prompts
- Feedback
BehiSecc/vibesec
目录来源: Specialized Domains
来源文档: https://github.com/BehiSecc/VibeSec-Skill/blob/main/SKILL.md
功能简介: 通过从漏洞猎人的视角来编写代码,帮助编写安全的代码,防止包括IDOR、XSS、SQL注入、SSRF和弱密码在内的常见漏洞。
适用时机: 当任务需要此目录条目中描述的能力时使用:通过从漏洞猎人的视角看待代码,通过防止包括 IDOR、XSS、SQL 注入、SSRF 和弱式身份验证在内的常见漏洞,有助于编写安全代码。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Secure Coding Guide for Web Applications
- Overview
- Access Control Issues
- Core Requirements
- Authorization Checks Checklist
- Common Pitfalls to Avoid
- Implementation Pattern
- Client-Side Bugs
- Cross-Site Scripting (XSS)
- Cross-Site Request Forgery (CSRF)
- Secret Keys and Sensitive Data Exposure
- Open Redirect
- Protection Strategies
- Bypass Techniques to Block
- IDN Homograph Attack Protection
- Password Security
better-auth/best-practices
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/blob/main/better-auth/best-practices/SKILL.md
功能简介: 更好的认证集成的最佳实践
适用时机: 当用户提到Better Auth、betterauth、auth.ts,或者需要使用电子邮件/密码、OAuth或插件配置来设置TypeScript认证时使用。
来源技能章节菜单:
- Better Auth Integration Guide
- Documentation Version
- Setup Workflow
- Quick Reference
- Environment Variables
- File Location
- CLI Commands
- Core Config Options
- Database
- Session Management
- User & Account Config
- Email Flows
- Security
- Hooks
- Plugins
- Client
better-auth/create-auth
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/blob/main/better-auth/create-auth/SKILL.md
功能简介: 使用Better Auth创建身份验证设置
适用时机: 当用户想使用Better Auth为新项目或现有项目添加登录、注册或身份验证时使用。
来源技能章节菜单:
- Create Auth Skill
- Phase 1: Planning (REQUIRED before implementation)
- Step 1: Scan the project
- Step 2: Ask planning questions
- Step 3: Summarize the plan
- Phase 2: Implementation
- Installation
- Environment Variables
- Server Config (auth.ts)
- Client Config (auth-client.ts)
- Route Handler Setup
- Database Migrations
- Database Adapters
- Drizzle + PostgreSQL Setup
- Drizzle Config (`drizzle.config.ts`)
- Common Plugins
better-auth/emailAndPassword
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/blob/main/better-auth/emailAndPassword/SKILL.md
功能简介: 使用Better Auth进行电子邮件和密码认证
适用时机: 当用户需要使用Better Auth设置登录、登录、注册、凭证认证或密码安全时使用。
来源技能章节菜单:
- Quick Start
- Email Verification Setup
- Requiring Email Verification
- Client Side Validation
- Callback URLs
- Password Reset Flows
- Security Considerations
- Sending the Password Reset
- Password Hashing
- Custom Hashing Algorithm
better-auth/explain-error
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/tree/main/better-auth/explain-error
功能简介: 更好地解释“认证”错误消息
适用时机: 查找 Better Auth 中 INVALID_SESSION 错误含义 在 Next.js 路由中为 Better Auth 的认证失败添加适当的捕获块 在调试 RATE_LIMIT_EXCEEDED 问题时查找相关错误代码 为 Better Auth 的 OAuth 回调错误生成生产级错误处理 了解何时可以重试 Better Auth 错误,何时是致命错误
未能取得单独的 SKILL.md;本文根据目录简介和可访问的仓库 README 整理。
better-auth/organization
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/blob/main/better-auth/organization/SKILL.md
功能简介: 使用Better Auth进行组织管理
适用时机: 当用户需要组织设置、团队管理、成员角色、访问控制或Better Auth组织插件时使用。
来源技能章节菜单:
- Setup
- Client-Side Setup
- Creating Organizations
- Controlling Organization Creation
- Creating Organizations on Behalf of Users
- Active Organizations
- Members
- Adding Members (Server-Side)
- Assigning Multiple Roles
- Removing Members
- Updating Member Roles
- Membership Limits
- Invitations
- Setting Up Invitation Emails
- Sending Invitations
- Shareable Invitation URLs
better-auth/providers
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/tree/main/better-auth/providers
功能简介: 更好的Auth认证提供商
适用时机: 查找Better Auth中GitHub OAuth所需的环境变量 检查Better Auth为Google登录所期望的回调URL格式 比较Discord和Twitter OAuth提供商的设置步骤 在编写认证路由的同时生成特定提供商的集成代码 审查无密码认证设置的特殊要求
未能取得单独的 SKILL.md;本文根据目录简介和可访问的仓库 README 整理。
better-auth/twoFactor
目录来源: Skills by Better Auth Team
来源文档: https://github.com/better-auth/skills/blob/main/better-auth/twoFactor/SKILL.md
功能简介: 使用Better Auth的双重身份验证
适用时机: 当用户需要使用Better Auth进行多因素身份验证(MFA)、多重身份验证、身份验证器设置或登录安全时使用。
来源技能章节菜单:
- Setup
- Client-Side Setup
- Enabling 2FA for Users
- TOTP (Authenticator App)
- Displaying the QR Code
- Verifying TOTP Codes
- TOTP Configuration Options
- OTP (Email/SMS)
- Configuring OTP Delivery
- Sending and Verifying OTP
- OTP Storage Security
- Backup Codes
- Displaying Backup Codes
- Regenerating Backup Codes
- Using Backup Codes for Recovery
- Backup Code Configuration
binance/query-token-audit
目录来源: Skills by Binance
来源文档: https://github.com/binance/binance-skills-hub/blob/main/skills/binance-web3/query-token-audit/SKILL.md
功能简介: 审计代币安全,以检测BSC、Base、Solana和以太坊上的诈骗、蜜罐和恶意合约
适用时机: 在用户询问“这个代币安全吗?”、“检查代币安全性”、“审计代币”或在任何兑换之前使用。
来源技能章节菜单:
- Query Token Audit Skill
- Overview
- Use Cases
- Supported Chains
- API: Token Security Audit
- Method: POST
- User Agent Header
- Notes
coinbase/authenticate-wallet
目录来源: Skills by Coinbase
来源文档: https://github.com/coinbase/agentic-wallet-skills/blob/main/README.md
功能简介: 通过电子邮件一次性密码处理Coinbase支付钱包的登录
适用时机: 在发送USDC之前登录您的Coinbase钱包 在交易过程中会话过期后重新验证 设置新机器上的钱包访问权限(首次) 在运行余额查询之前检查钱包认证状态 在交易期间恢复“未登录”错误
来源技能章节菜单:
- Coinbase Agentic Wallet Skills
- What it does
- Installation
- Usage
- Contributing
- Updating the `awal` version
- License
composiohq/composio
目录来源: Skills by Composio Team
来源文档: https://github.com/ComposioHQ/skills/blob/main/skills/composio/SKILL.md
功能简介: 使用管理式身份验证将AI代理连接到1000多个外部应用程序
适用时机: 从终端命令发送 Gmail 草稿,无需编写任何代码 将多租户 AI 代理连接到每个用户的专属 Slack 工作区 根据使用场景搜索 1000 多种工具集成,以找到合适的操作 在对话中触发 AI 代理创建 GitHub 问题 监听来自外部应用的实时事件,以驱动代理工作流程
来源技能章节菜单:
- When to Apply
- Setup
- 1. Use Apps via Composio CLI
- 2. Building Apps and Agents with Composio
danoszz/penthera
目录来源: n8n Automation
来源文档: https://github.com/danoszz/penthera/blob/main/skills/penthera/SKILL.md
功能简介: 针对仓库、依赖项和代理配置的离线基于规则的安全检查
适用时机: 当用户请求扫描、审计、加固、修复或审查其自身应用的安全性、测试环境URL、本地主机、仓库、依赖项或代理设置时使用。未经明确授权,请勿用于目标。
来源技能章节菜单:
- Penthera Security Scanner
- Critical: Authorization gate
- Preflight
- Resolve CLI command
- Decision tree
- Default safe behavior
- Reply with the brief
- Destructive mode gate
- Workflow 1: Pre-release audit
- Workflow 2: Repo + live combined
- Workflow 3: CI / baseline regression
- Workflow 4: Scan and fix
- Fix-mode gate
- Workflow 5: Owner reports & questionnaire responses
- Authenticated scans
- Workflow 6: Agent config audit
firebase/firebase-auth-basics
目录来源: Skills by Firebase
来源文档: https://github.com/firebase/agent-skills/blob/main/skills/firebase-auth-basics/SKILL.md
功能简介: 使用登录提供商设置Firebase身份验证
适用时机: 当用户的 App 需要使用身份验证规则进行用户登录、用户管理或安全的数据访问时,请使用此技能。
来源技能章节菜单:
- Prerequisites
- Core Concepts
- Users
- Identity Providers
- Tokens
- Workflow
- 1. Provisioning
- 2. Client Setup & Usage
- 3. Security Rules
firebase/firebase-basics
目录来源: Skills by Firebase
来源文档: https://github.com/firebase/agent-skills/blob/main/skills/firebase-basics/SKILL.md
功能简介: 处理Firebase CLI的安装、身份验证和日常工作流程
适用时机: 在终端中从零开始设置一个新的Firebase项目 在没有浏览器的情况下在远程服务器上登录Firebase 在开发过程中切换多个Firebase项目 更新后修复损坏的Firebase CLI安装 添加像Hosting或Firestore这样的新Firebase服务到现有项目中
来源技能章节菜单:
- Prerequisites
- Firebase Usage Principles
- References
- Common Issues
firebase/firebase-security-rules-auditor
目录来源: Skills by Firebase
来源文档: https://github.com/firebase/agent-skills/blob/main/skills/firebase-security-rules-auditor/SKILL.md
功能简介: 审计Firestore安全规则并标记有风险的模式
适用时机: 用于审计/审查规则、运行红队规则评估或根据审计员检查清单进行评分。不要用于 Firebase CLI(登录、部署)、Auth、Crashlytics、远程配置或数据库查询。
来源技能章节菜单:
- Overview
- Scoring Criteria
- Assessment: Security Validator (Red Team Edition)
- Mandatory Audit Checklist:
- Admin Bootstrapping & Privileges:
- Scoring Criteria (1-5):
firecrawl/firecrawl-build-interact
目录来源: Skills by Firecrawl Team
来源文档: https://github.com/firecrawl/skills/blob/main/README.md
功能简介: 多步骤Firecrawl浏览器流程:点击、填写表单、分页和考虑身份验证的导航
适用时机: 登录仪表盘并爬取认证后的数据 通过分页搜索结果点击以收集条目 从应用代码中填写并提交多步骤网页表单 驱动无限滚动动态以捕获所有项目 导航依赖JS的单页应用,在交互时显示内容
来源技能章节菜单:
- Firecrawl Skills
- Install
- Catalog layout
- Contributing: where does my PR go?
- Skills
- Core (`skills/core/`)
- Build (`skills/build/`)
- Workflows (`skills/workflows/`)
- MCP Server
- Plugins
- Prerequisites
- Which skills do I need?
- Docs (Source of Truth)
- License
garrytan/cso
目录来源: Skills by Garry Tan (gstack)
来源文档: https://github.com/garrytan/gstack/blob/main/cso/SKILL.md
功能简介: 首席安全官:OWASP前10+STRIDE威胁模型,无假阳性排除
适用时机: 在仓库公开前扫描 git 历史记录以查找泄露的 AWS 密钥和 OAuth 令牌 审计 GitHub Actions 工作流中的 pull_request_target 滥用行为和未挂钩的第三方动作 检查 Webhook 处理程序,确保整个代码库中均有 HMAC 签名验证 运行 LLM 功能审计,以追踪到达系统提示的用户输入路径 在进行 SOC 2 或合规审查前的审计周期中跟踪安全态势趋势
来源技能章节菜单:
- /cso — evidence before assurance
- Arguments
- Mode Resolution
- Section index — Read each section when its situation applies
- Phase 0: Application model and readiness
- Phase 1: Attack surface census
- Phase 12: Evidence rubric and independent challenge
- Phase 13: Report and repair artifacts
- Phase 14: Recovery and current-source rechecks
google/cloud/google-cloud-recipe-auth
目录来源: Skills by Google Cloud
来源文档: https://github.com/google/skills/blob/main/skills/cloud/google-cloud-recipe-auth/SKILL.md
功能简介: 提供关于验证和授权Google云服务和API的专家指导,涵盖人类用户、服务身份、应用程序默认凭证(ADC)和b...
适用时机: 当任务需要此目录条目中描述的能力时使用:提供有关验证和授权Google云服务和API的专家指导,涵盖人类用户、服务身份、应用程序默认凭证(ADC)和b...
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Authenticating and Authorizing to Google Cloud (IAM)
- Attribution
- Authentication
- Clarifying Questions for the Agent
- Human Authentication
- Types of User Identities
- Methods of Access for Developers and Administrators
- For End-Users and Customers
- Service-to-Service Authentication
- Service Accounts and Service Agents
- Best Practice: Attaching Service Accounts
- Special Cases & Advanced Topics
- Authorization
- Examples
- Human-to-Service (Local Python Development)
- Service-to-Service (Cloud Run to Cloud SQL)
google/cloud/google-cloud-waf-security
目录来源: Skills by Google Cloud
来源文档: https://github.com/google/skills/blob/main/skills/cloud/google-cloud-waf-security/SKILL.md
功能简介: 基于Google Cloud Well-Architected Framework (WAF)中的设计原则和建议,为Google Cloud工作负载生成以安全为重点的指导。
适用时机: 当任务需要此目录条目中描述的能力时使用:基于Google Cloud Well-Architected Framework (WAF)中的设计原则和建议,为Google Cloud工作负载生成以安全为重点的指导。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Google Cloud Well-Architected Framework skill for the Security pillar
- Attribution
- Overview
- Workflow
- Core principles
- Relevant Google Cloud products
- Workload assessment questions
- Validation checklist
mcollina/skills
目录来源: Development and Testing
来源文档: https://github.com/mcollina/skills/blob/main/README.md
功能简介: 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、技能优化器等
适用时机: 当任务需要此目录条目中描述的能力时使用: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、技能优化器等
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Matteo Collina’s Skills
- Available Skills
- Usage
- Benchmarking
此目录条目指向一个技能集合。扫描来源仓库发现 11 个不重复的技能文件路径。
部分成员: documentation · fastify · init · linting-neostandard-eslint9 · node · nodejs-core · oauth · octocat
更多成员请查看来源仓库。
microsoft/azure-identity-dotnet
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-dotnet/skills/azure-identity-dotnet/SKILL.md
功能简介: Microsoft Entra ID 身份验证
适用时机: 在不硬编码密钥的情况下验证 Azure SDK 客户端(Blob 存储、密钥宝) 从本地的 AzureCliCredential 切换到生产环境中的 ManagedIdentityCredential,无需代码更改 使用客户端 ID 在 AKS 工作负载上配置用户指定的托管身份 使用 ChainedTokenCredential 与备用凭据串联,实现多环境部署 对 Azure 政府或中国主权云端点进行身份验证
来源技能章节菜单:
- Azure Identity library for .NET
- Installation
- Environment Variables
- Service Principal with Secret
- Service Principal with Certificate
- Managed Identity
- DefaultAzureCredential
- Basic Usage
- ASP.NET Core with Dependency Injection
- Customizing DefaultAzureCredential
- Credential Types
- ManagedIdentityCredential (Production)
- ClientSecretCredential
- ClientCertificateCredential
- ChainedTokenCredential (Custom Chain)
- Developer Credentials
microsoft/azure-identity-java
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-java/skills/azure-identity-java/SKILL.md
功能简介: Microsoft Entra ID 身份验证
适用时机: 在Java应用程序中实现默认AzureCredential、托管身份、服务主密钥或任何Azure身份验证模式时使用。
来源技能章节菜单:
- Azure Identity library for Java
- Installation
- Key Concepts
- DefaultAzureCredential (Recommended)
- Configure DefaultAzureCredential
- Managed Identity
- Service Principal with Secret
- Service Principal with Certificate
- Environment Credential
- Required Environment Variables
- Azure CLI Credential
- Interactive Browser
- Device Code
- Chained Credential
- Workload Identity (AKS)
- Token Caching
microsoft/azure-identity-py
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-python/skills/azure-identity-py/SKILL.md
功能简介: Microsoft Entra ID 身份验证
适用时机: 使用Python在Azure Blob存储客户端中进行身份验证,无需管理凭证 从本地Azure CLI登录切换到生产环境中管理的身份,且无需更改代码 通过OpenAI SDK的get_bearer_token_provider获取Azure OpenAI的Bearer令牌 为在AKS上运行的Python应用程序设置工作负载身份认证 将ManagedIdentityCredential与AzureCliCredential串联,作为开发/生产备用方案
来源技能章节菜单:
- Azure Identity library for Python
- Installation
- Python Version
- Environment Variables
- Authentication & Lifecycle
- DefaultAzureCredential
- Credential Chain Order
- Customizing DefaultAzureCredential
- Exclude Parameters
- get_bearer_token_provider
- Credential Types
- Credential Chains
- Azure-Hosted Applications
- Service Principals
- User Authentication
- Developer Tools
microsoft/azure-identity-rust
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-rust/skills/azure-identity-rust/SKILL.md
功能简介: Microsoft Entra ID 身份验证
适用时机: 使用Azure CLI凭据在本地验证Rust应用程序与Azure密钥库的连接 从AKS工作集(带有工作负载身份)连接到Azure Blob存储 在CI中旋转服务主密钥,而无需更改应用程序代码 使用Azure Functions上的系统分配管理身份访问Azure SQL 在生产环境中无需代码更改,从本地DeveloperToolsCredential切换到ManagedIdentityCredential
来源技能章节菜单:
- Azure Identity library for Rust
- Installation
- Environment Variables
- Authentication
- DeveloperToolsCredential (Local Development)
- ManagedIdentityCredential (Production)
- ClientSecretCredential (Service Principal)
- Credential Types
- Best Practices
- Reference Links
microsoft/azure-identity-ts
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-typescript/skills/azure-identity-ts/SKILL.md
功能简介: Microsoft Entra ID 身份验证
适用时机: 在使用默认AzureCredential、管理身份、服务主导者或交互式浏览器登录进行身份验证时使用。
来源技能章节菜单:
- Azure Identity library for TypeScript
- Installation
- Environment Variables
- Service Principal (Secret)
- Service Principal (Certificate)
- Workload Identity (Kubernetes)
- DefaultAzureCredential (Recommended for Local Development)
- Managed Identity
- System-Assigned
- User-Assigned (by Client ID)
- User-Assigned (by Resource ID)
- User-Assigned (by Object ID)
- Service Principal
- Client Secret
- Client Certificate
- Interactive Authentication
microsoft/azure-keyvault-py
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-python/skills/azure-keyvault-py/SKILL.md
功能简介: 秘密、密钥和证书管理
适用时机: 在运行时从数据库中获取凭据,而不是将其硬编码在配置文件中 通过更新单个安全库密钥,将 API 密钥在多个服务之间轮换 使用存储在安全库中的 RSA 密钥签名数据包,而不会暴露私钥 管理 Azure 托管应用的 TLS 证书生命周期 在写入存储之前,使用由安全库管理的密钥加密敏感的用户数据
来源技能章节菜单:
- Azure Key Vault SDK for Python
- Installation
- Environment Variables
- Authentication & Lifecycle
- Secrets
- SecretClient Setup
- Secret Operations
- Keys
- KeyClient Setup
- Key Operations
- Cryptographic Operations
- Certificates
- CertificateClient Setup
- Certificate Operations
- Client Types Table
- Async Clients
microsoft/azure-keyvault-secrets-rust
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-rust/skills/azure-keyvault-secrets-rust/SKILL.md
功能简介: 密钥宝秘密存储
适用时机: 在应用程序启动时从密钥库读取数据库密码,而不是硬编码它们。在不重新部署Rust服务的情况下,旋转存储在密钥库中的API密钥。按环境对秘密进行标记,以分离生产和测试环境的凭据。获取特定秘密版本,以审计在特定日期上有效的数值。列出密钥库中的所有秘密,以在项目内进行凭据库存管理。
来源技能章节菜单:
- Azure Key Vault Secrets library for Rust
- Installation
- Environment Variables
- Authentication
- Core Workflow
- Set Secret
- Update Secret Properties
- Delete Secret
- List Secrets (Pagination)
- Error Handling
- RBAC Roles
- Best Practices
- Reference Links
microsoft/azure-keyvault-secrets-ts
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-typescript/skills/azure-keyvault-secrets-ts/SKILL.md
功能简介: 秘密存储和检索
适用时机: 用于存储和检索应用程序秘密或配置值时使用。
来源技能章节菜单:
- Azure Key Vault Secrets SDK for TypeScript
- Installation
- Environment Variables
- Authentication
- Secrets Operations
- Create/Set Secret
- Get Secret
- List Secrets
- Delete Secret
- Keys Operations
- Create Keys
- Get Key
- List Keys
- Rotate Key
- Delete Key
- Cryptographic Operations
microsoft/azure-security-keyvault-keys-dotnet
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-dotnet/skills/azure-security-keyvault-keys-dotnet/SKILL.md
功能简介: 密码学密钥管理
适用时机: 按90天周期轮换RSA密钥,并采用自动密钥库策略 使用RSA-OAEP加密AES会话密钥,然后再将其存储在数据库中 使用由托管式HSM支持的EC密钥签署JWT或数据包 使用数据加密密钥进行包装,确保明文密钥材料永远不会离开密钥库 列出所有密钥版本,以审计生产环境中哪些版本处于活跃状态
来源技能章节菜单:
- Azure.Security.KeyVault.Keys (.NET)
- Installation
- Environment Variables
- Client Hierarchy
- Authentication
- Microsoft Entra Token Credential
- Service Principal
- Key Management
- Create Keys
- Retrieve Keys
- Update Key Properties
- Delete and Purge Keys
- Backup and Restore
- Cryptographic Operations
- Get CryptographyClient
- Encrypt and Decrypt
microsoft/azure-security-keyvault-keys-java
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-java/skills/azure-security-keyvault-keys-java/SKILL.md
功能简介: 密码学密钥管理
适用时机: 在创建、管理或使用RSA/EC密钥、执行加密/解密/签名/验证操作或与HSM支持的密钥一起工作时使用。
来源技能章节菜单:
- Azure Key Vault Keys (Java)
- Installation
- Client Creation
- Key Types
- Create Keys
- Create RSA Key
- Create EC Key
- Create Symmetric Key (Managed HSM only)
- Get Key
- Update Key Properties
- List Keys
- Delete Key
- Cryptographic Operations
- Encrypt/Decrypt
- Sign/Verify
- Wrap/Unwrap Key
microsoft/azure-security-keyvault-secrets-java
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-java/skills/azure-security-keyvault-secrets-java/SKILL.md
功能简介: 密码和密钥的秘密管理
适用时机: 在存储、检索或管理密码、API密钥、连接字符串或其他敏感的配置数据时使用。
来源技能章节菜单:
- Azure Key Vault Secrets (Java)
- Installation
- Client Creation
- Create/Set Secret
- Get Secret
- Update Secret Properties
- List Secrets
- Delete Secret
- Recover Deleted Secret
- Purge Deleted Secret
- Backup and Restore
- Async Operations
- Configuration Patterns
- Load Multiple Secrets
- Secret Rotation Pattern
- Error Handling
microsoft/fastapi-router-py
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-python/skills/fastapi-router-py/SKILL.md
功能简介: 具有CRUD和身份验证的FastAPI路由器
适用时机: 在构建REST API端点、创建新路由、实现CRUD操作或在FastAPI应用程序中添加已认证端点时使用。
来源技能章节菜单:
- FastAPI Router
- Quick Start
- Authentication Patterns
- Response Models
- HTTP Status Codes
- Integration Steps
- Best Practices
- Reference Files
microsoft/microsoft-azure-webjobs-extensions-authentication-events-dotnet
目录来源: Skills by Microsoft
来源文档: https://github.com/microsoft/skills/blob/main/.github/plugins/azure-sdk-dotnet/skills/microsoft-azure-webjobs-extensions-authentication-events-dotnet/SKILL.md
功能简介: Entra ID自定义认证事件处理器
适用时机: 在 Entra 登录期间将部门和成本中心索赔添加到令牌中 根据用户的电子邮件域名预填充注册表字段 在属性提交时阻止来自特定电子邮件域的注册 通过自定义短信服务提供商发送 OTP 代码,而不是使用 Entra 默认设置 在将用户提交的属性保存到目录之前进行验证和规范化
来源技能章节菜单:
- Microsoft.Azure.WebJobs.Extensions.AuthenticationEvents (.NET)
- Installation
- Supported Events
- Core Workflows
- 1. Token Enrichment (Add Custom Claims)
- 2. Token Enrichment with External Data
- 3. Attribute Collection - Customize UI (Start Event)
- 4. Attribute Collection - Validate Submission (Submit Event)
- 5. Custom OTP Delivery
- 6. Function App Configuration
- Key Types Reference
- Entra ID Configuration
- Required App Registration Settings
- Best Practices
- Error Handling
- Related SDKs
MiniMax-AI/fullstack-dev
目录来源: Skills by MiniMax Team
来源文档: https://github.com/MiniMax-AI/skills/blob/main/skills/fullstack-dev/SKILL.md
功能简介: 采用REST API设计、身份验证流程、实时功能和数据库集成的前端架构
适用时机: 使用适当的服务层构建新的 Express 或 Django API 连接带有类型化 API 客户端和错误处理的 React 前端到带有 REST 后端的连接 为现有应用添加带自动令牌刷新功能的 JWT 身份验证 实现通过预签名的 S3 URL 上传文件,而不是通过服务器进行流式传输 设置用于实时通知或聊天的 SSE 或 WebSocket 端点
来源技能章节菜单:
- Full-Stack Development Practices
- MANDATORY WORKFLOW — Follow These Steps In Order
- Step 0: Gather Requirements
- Step 1: Architectural Decisions
- Step 2: Scaffold with Checklist
- Step 3: Implement Following Patterns
- Step 4: Test & Verify
- Step 5: Handoff Summary
- Scope
- Quick Start — New Backend Service Checklist
- Quick Start — Frontend-Backend Integration Checklist
- Quick Navigation
- Core Principles (7 Iron Rules)
- 1. Project Structure & Layering (CRITICAL)
- Feature-First Organization
- Three-Layer Architecture
MiniMax-AI/ios-application-dev
目录来源: Skills by MiniMax Team
来源文档: https://github.com/MiniMax-AI/skills/blob/main/skills/ios-application-dev/SKILL.md
功能简介: 使用UIKit、SnapKit和SwiftUI进行iOS开发,涵盖导航、暗模式和HIG合规性
适用时机: 使用时机:开发iOS应用程序、实现UI、审查iOS代码、与UIKit/SnapKit/SwiftUI布局合作、构建iPhone界面、Swift移动开发、符合Apple HIG、实现iOS无障碍功能。
来源技能章节菜单:
- iOS Application Development Guide
- Quick Reference
- UIKit
- SwiftUI
- Core Principles
- Layout
- Typography
- Colors
- Accessibility
- Navigation
- Privacy & Permissions
- Checklist
- Components
- Privacy
- System Integration
- References
mongodb/mongodb-mcp-setup
目录来源: Skills by MongoDB
来源文档: https://github.com/mongodb/agent-skills/blob/main/skills/mongodb-mcp-setup/SKILL.md
功能简介: 设置带有身份验证和连接配置的MongoDB MCP服务器
适用时机: 当用户已安装MongoDB MCP服务器但尚未配置所需的环境变量,或他们询问如何连接到MongoDB/Atlas但尚未设置凭据时,请使用此技能。
来源技能章节菜单:
- MongoDB MCP Server Setup
- Overview
- Step 0: Detect Client
- Step 1: Check Existing Configuration
- Step 2: Present Configuration Options
- Step 3a: Connection String Setup
- 3a.1: Explain How to Find the Connection String
- Step 3b: Service Account Setup
- 3b.1: Guide Through Atlas Service Account Creation
- 3b.2: API Access List Configuration
- Step 3c: Atlas Local Setup
- 3c.1: Check Docker Installation
- 3c.2: Confirm Setup Complete
- Step 4: Determine Read-Only vs Read-Write Access
- Step 5: Configure Credentials
- 5.1: Add credentials
moxno/privacyscrubber-guard
目录来源: Development and Testing
来源文档: https://github.com/moxno/privacyscrubber-mcp/blob/main/skills/privacyscrubber-guard/SKILL.md
功能简介: 在代理调用之前对内存中的PII和机密信息进行消毒处理
适用时机: 当任务需要此目录条目中描述的能力时使用:在代理调用之前对内存中的PII和秘密进行消毒处理
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- PrivacyScrubber Guard — Agent Skill
- Quick Setup
- Core Agent Workflow
mukul975/Anthropic-Cybersecurity-Skills
目录来源: Development and Testing
来源文档: https://github.com/mukul975/Anthropic-Cybersecurity-Skills/blob/main/README.md
功能简介: 753项涵盖38个领域的网络安全技能:云安全、渗透测试、红队、DFIR、恶意软件分析、威胁情报等(已映射到MITRE ATT&CK)
适用时机: 当任务需要本目录条目中描述的能力时使用:753项涵盖38个领域的网络安全技能:云安全、渗透测试、红队、DFIR、恶意软件分析、威胁情报等(已映射到MITRE ATT&CK)
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Anthropic Cybersecurity Skills
- The largest open-source cybersecurity skills library for AI agents
- Give any AI agent the security skills of a senior analyst
- Six frameworks, one skill library
- 🆕 MITRE Fight Fraud Framework (F3) — 94 fraud-relevant skills
- MITRE ATT&CK v19.1 — 805/817 skills mapped
- Quick start
- 🌍 GARS-2026 — Global Agentic AI Readiness Survey
- 🚀 Try it on the Playground
- Why this exists
- What’s inside — 34 security domains
- How AI agents use these skills
- Skill anatomy
- YAML frontmatter (real example)
- Markdown body sections
- MITRE ATLAS 2026.07 — AI/ML adversarial threats
此目录条目指向一个技能集合。扫描来源仓库发现 818 个不重复的技能文件路径。
部分成员: abusing-dpapi-for-credential-access · abusing-shadow-credentials-for-privesc · achieving-cmmc-level-2-compliance · acquiring-disk-image-with-dd-and-dcfldd · analyzing-active-directory-acl-abuse · analyzing-android-malware-with-apktool · analyzing-api-gateway-access-logs · analyzing-apt-group-with-mitre-navigator
更多成员请查看来源仓库。
NeoLabHQ/review
目录来源: Development and Testing
来源文档: https://github.com/NeoLabHQ/context-engineering-kit/blob/master/plugins/review/README.md
功能简介: 使用专业代理人进行全面的PR代码审查:bug-hunter、security-auditor、code-quality-reviewer、contracts-reviewer、historical-context-reviewer、test-coverage-reviewer
适用时机: 当任务需要本目录条目中描述的能力时使用:使用专门的代理人进行全面的PR代码审查:bug-hunter、security-auditor、code-quality-reviewer、contracts-reviewer、historical-context-reviewer、test-coverage-reviewer
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Review Plugin
- Focused on
- Overview
- Quick Start
- Advanced Usage
- CI/CD Integration
- Agent Architecture
- Commands
- Review Agents
- Bug Hunter
- Security Auditor
- Test Coverage Reviewer
- Code Quality Reviewer
- Contracts Reviewer
- Historical Context Reviewer
- Change Triage Agents
此目录条目指向一个技能集合。扫描来源仓库发现 3 个不重复的技能文件路径。
部分成员: review-local-changes · review-pr · traiage-review
openai/netlify-deploy
目录来源: Skills by OpenAI
来源文档: https://github.com/openai/skills/blob/main/skills/.curated/netlify-deploy/SKILL.md
功能简介: 使用CLI认证、链接和环境支持自动化Netlify部署
适用时机: 当用户要求在Netlify上部署、托管、发布或链接一个网站/仓库时使用,包括预览和生产部署。
来源技能章节菜单:
- Netlify Deployment Skill
- Overview
- Prerequisites
- Authentication Pattern
- Workflow
- 1. Verify Netlify CLI Authentication
- 2. Detect Site Link Status
- 3. Link to Existing Site or Create New
- 4. Verify Dependencies
- 5. Deploy to Netlify
- 6. Report Results
- Handling netlify.toml
- Example Full Workflow
- Error Handling
- Troubleshooting
- Escalated Network Access
openai/security-best-practices
目录来源: Skills by OpenAI
来源文档: https://github.com/openai/skills/blob/main/skills/.curated/security-best-practices/SKILL.md
功能简介: 审查针对特定语言的安全漏洞代码
适用时机: 在生产部署前对 Flask API 进行常见 Python 网页安全问题的审计 为 Node.js Express 应用生成等级排序的漏洞报告 使用已应用安全默认值的全新 Go HTTP 处理程序 在现有 TypeScript 项目中添加功能时捕获硬编码的密钥或缺失的输入验证 获取与相关安全指南相关的解释的特定 CVE 类问题修复方案
来源技能章节菜单:
- Security Best Practices
- Overview
- Workflow
- Workflow Decision Tree
- Overrides
- Report Format
- Fixes
- General Security Advice
- Avoid Using Incrementing IDs for Public IDs of Resources
- A note on TLS
openai/security-ownership-map
目录来源: Skills by OpenAI
来源文档: https://github.com/openai/skills/blob/main/skills/.curated/security-ownership-map/SKILL.md
功能简介: 绘制人与文件所有权图,计算乘数因子,并识别风险
适用时机: 使用1的bus factor查找认证或加密文件 检查实际拥有敏感代码的人员,与CODEOWNERS所列的人员进行对比 定位数月无人触碰的过时的安全关键文件 按共同更改模式对文件进行分组,以发现隐藏的所有权集群 跟踪哪些贡献者拥有最高敏感更改的集中度
来源技能章节菜单:
- Security Ownership Map
- Overview
- Requirements
- Workflow
- Quick start
- Sensitivity rules
- Output artifacts
- LLM query helper
- Basic security queries
- Summary format (default)
- Graph persistence
- Notes
openai/security-threat-model
目录来源: Skills by OpenAI
来源文档: https://github.com/openai/skills/blob/main/skills/.curated/security-threat-model/SKILL.md
功能简介: 生成特定仓库的威胁模型,识别信任边界
适用时机: 在安全审查前对网络API进行威胁建模 在多服务后端中映射信任边界 识别文件上传或解析器端点的滥用路径 为渗透测试范围生成威胁模型文档 对面向外暴露的管理员接口进行攻击者能力审计
来源技能章节菜单:
- Threat Model Source Code Repo
- Quick start
- Workflow
- 1) Scope and extract the system model
- 2) Derive boundaries, assets, and entry points
- 3) Calibrate assets and attacker capabilities
- 4) Enumerate threats as abuse paths
- 5) Prioritize with explicit likelihood and impact reasoning
- 6) Validate service context and assumptions with the user
- 7) Recommend mitigations and focus paths
- 8) Run a quality check before finalizing
- Risk prioritization guidance (illustrative, not exhaustive)
- References
prompt-security/clawsec
目录来源: Specialized Domains
来源文档: https://github.com/prompt-security/clawsec/blob/main/README.md
功能简介: 包含漂移检测、自动审计和技能完整性验证的安全技能套件
适用时机: 当任务需要本目录条目中描述的能力时使用:包含漂移检测、自动审计和技能完整性验证的安全技能套件
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Install the OpenClaw suite
- 1. Add the suite
- 2. Review and enable the advisory hook
- See it work
- Detect and respond to agent-file drift
- How ClawSec protects an agent
- Platform entry points
- Skill feature matrix
- Query the signed advisory channel
- Build, test, and contribute
- Translations
- License
此目录条目指向一个技能集合。扫描来源仓库发现 16 个不重复的技能文件路径。
部分成员: claw-release · clawsec-clawhub-checker · clawsec-feed · clawsec-nanoclaw · clawsec-scanner · clawsec-suite · clawtributor · hermes-attestation-guardian
更多成员请查看来源仓库。
redhat/cve-skillpack
目录来源: Skills by RedHat
来源文档: https://catalog.redhat.com/en/ai/skills/detail/agentic-skill-pack-for-red-hat-customers#agent-and-skills
功能简介: 了解CVE,检查产品生命周期状态,收集诊断信息,并按适当的严重程度提交支持案例——这是日常运营中必不可少的Red Hat技能。
适用时机: 当任务需要本目录条目中描述的能力时使用:了解CVE,检查产品生命周期状态,收集诊断信息,并在适当的严重程度下提交支持案例——这是日常运营中必不可少的Red Hat技能。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Platforms
- Products & services
- Try, buy, sell
- Help
- About Red Hat Ecosystem Catalog
- Red Hat legal and privacy links
- Red Hat legal and privacy links
此目录条目指向一个技能集合。扫描来源仓库发现 0 个不重复的技能文件路径。
未能取得单独的 SKILL.md;本文根据目录简介和可访问的仓库 README 整理。
Rootly-AI-Labs/rootly-incident-responder
目录来源: Development and Testing
来源文档: https://github.com/rootlyhq/rootly-mcp-server/blob/main/examples/skills/rootly-incident-responder.md
功能简介: 采用机器学习相似匹配、解决方案建议和值班协调的AI驱动事件响应。需要[Rootly MCP Server](https://github\.com/rootlyhq/rootly\-mcp\-server\)
适用时机: 当任务需要本目录条目中描述的能力时使用:采用AI驱动的事件响应,支持机器学习相似匹配、解决方案建议和值班协调。需要[Rootly MCP Server](https://github\.com/rootlyhq/rootly\-mcp\-server\)
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Rootly Incident Responder
- Core Principles
- Your Workflow
- 1. Gather Comprehensive Incident Context
- 2. Find Historical Context
- 3. Get Intelligent Solution Recommendations
- 4. Identify On-Call Team & Stakeholders
- 5. Correlate with Code Changes
- 6. Analyze Root Cause
- 7. Create Action Items & Remediation Plan
- 8. Document Resolution
- Best Practices
- Prioritization
- Communication
- Uncertainty Handling
- Leveraging Rootly’s Intelligence
saleh-alhaddad/itqan-engineering
目录来源: Development and Testing
来源文档: https://github.com/saleh-alhaddad/itqan-engineering/blob/main/SKILL.md
功能简介: 12项技能涵盖完整的软件工程生命周期:一个可供简历撰写的协调器,包含规范、计划、TDD构建、验证、五轴审查、安全与发布——代码前设有审批关卡,“完成”前需提供证据。
适用时机: 当任务需要本目录条目中描述的能力时使用:12项技能中的完整软件工程生命周期:一个可复现的调配器,包含规范、计划、TDD构建、验证、五轴审查、安全和发布——代码前有批准门槛,“完成”前有证据。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- itqan — engineering skills suite router
- How to work
- Non-negotiables (from CONVENTIONS)
此目录条目指向一个技能集合。扫描来源仓库发现 0 个不重复的技能文件路径。
SHADOWPR0/security-bluebook-builder
目录来源: Specialized Domains
来源文档: https://github.com/SHADOWPR0/security-bluebook-builder/blob/main/SKILL.md
功能简介: 为敏感应用程序构建安全蓝皮书
适用时机: 当用户需要针对处理PII/PHI/财务数据的应用程序的威胁模型、数据分类规则、身份验证/会话政策、日志记录和审计要求、保留/删除期望、事件响应或安全门槛时使用。
来源技能章节菜单:
- Security Bluebook Builder
- Overview
- Workflow
- 1) Gather inputs (ask only if missing)
- 2) Draft the Blue Book
- 3) Enforce guardrails
- 4) Quality checks
- Resources
squirrelscan/squirrelscan
目录来源: Development and Testing
来源文档: https://github.com/squirrelscan/skills/blob/main/skills/squirrelscan/SKILL.md
功能简介: 对网站进行SEO、性能、安全、可访问性和返现修复的审计
适用时机: 当用户希望检查、审计或改进网站的SEO、排名、速度或健康状况,以及用于squirrelscan本身的所有操作(如安装或更新CLI、登录和API密钥、运行审计、发布和共享报告、云积分、MCP服务器设置、配置或故障排除)时使用。还涵盖实体地图:网站在其JSON-LD中声明的实体的全站级图表,以及修复结构化数据身份问题,例如在每页单独声明的组织。
来源技能章节菜单:
- squirrelscan CLI
- Links
- Install
- Command overview
- Quickstart
- Exit codes and audits that did not run
- Audit levels
- Probing intensity
- Authentication and accounts
- Reports
- Publishing
- Regression diffs
- Skipped checks
- Copy fix prompts
- Not checked
- What a published report holds
Tencent/aig-agent-redteam
目录来源: Specialized Domains
来源文档: https://github.com/Tencent/AI-Infra-Guard/blob/main/skills/aig-agent-redteam/SKILL.md
功能简介: 单命令代理红队安全评估技能
适用时机: 当任务需要本目录条目中描述的能力时使用:一个命令代理红队安全评估技能
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- AIG Agent 蓝军安全演习
- 操作原则
- 最小脚本策略
- 复用 AI-Infra-Guard 数据源
- 演习流程
- 报告资源
- Step 0:范围与安全边界
- Step 1:能力与信任边界建模
- Step 2:生成风险假设
- Step 3:规划无害测试
- Step 4:执行并自适应变异
- Step 5:静态代码与供应链审计
- Step 6:基础设施审查
- Step 7:证据、评级与报告
- 模块指引
- 报告模板
trailofbits/ask-questions-if-underspecified
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/README.md
功能简介: 关于模糊要求进行澄清的提示
适用时机: 在可能影响数十个文件的重构之前询问目标文件范围 在升级依赖项之前获取兼容性要求的确认 在为规格不详的功能编写测试之前明确接受标准 在运行数据迁移之前解决模糊的撤销要求 在性能优化通过之前确认哪些组件在范围之内
来源技能章节菜单:
- Trail of Bits Skills Marketplace
- Installation
- Claude Code Marketplace
- Browse and Install Plugins
- Codex
- ChatGPT Workspace Marketplace
- Local Development
- Available Plugins
- Smart Contract Security
- Code Auditing
- Malware Analysis
- Verification
- Reverse Engineering
- Mobile Security
- Development
- Team Management
trailofbits/audit-context-building
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/audit-context-building/skills/audit-context-building/SKILL.md
功能简介: 通过超粒度代码分析获得深层次的建筑语境
适用时机: 在对不熟悉的代码进行审计、威胁模型或架构审查时使用,并在任何漏洞挖掘通过之前使用。
来源技能章节菜单:
- Audit Context Building
- When to Use
- When NOT to Use
- Do not analyze in this context
- What comes back, and how to read it
- The format
trailofbits/building-secure-contracts
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/building-secure-contracts/README.md
功能简介: 包含6个区块链漏洞扫描器的智能合约安全工具包
适用时机: 扫描 Solana/Anchor 应用程序以检查缺失的所有权检查和签名者授权问题 检查 ERC20 代币实现是否存在如转账费用或重新基准等非标准行为 准备智能合约代码库以配合第三方安全审计,并采用结构化检查清单 评估 Substrate 托盘在处理 BadOrigin 和验证未签名交易方面的能力 在大规模部署前对合同的 9 个类别进行代码成熟度评分
来源技能章节菜单:
- Building Secure Contracts
- Overview
- Installation
- Vulnerability Scanners
- Algorand Vulnerability Scanner
- Cairo Vulnerability Scanner
- Cosmos Vulnerability Scanner
- Solana Vulnerability Scanner
- Substrate Vulnerability Scanner
- TON Vulnerability Scanner
- Development Guidelines Assistants
- Audit Prep Assistant
- Code Maturity Assessor
- Guidelines Advisor
- Secure Workflow Guide
- Token Integration Analyzer
此目录条目指向一个技能集合。扫描来源仓库发现 11 个不重复的技能文件路径。
部分成员: algorand-vulnerability-scanner · audit-prep-assistant · cairo-vulnerability-scanner · code-maturity-assessor · cosmos-vulnerability-scanner · guidelines-advisor · secure-workflow-guide · solana-vulnerability-scanner
更多成员请查看来源仓库。
trailofbits/burpsuite-project-parser
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/burpsuite-project-parser/skills/burpsuite-project-parser/SKILL.md
功能简介: 从Burp Suite项目文件中搜索和提取数据
适用时机: 使用该工具在使用正则表达式模式搜索响应头或体时,提取安全审计发现、导出代理历史记录或网站地图数据,或分析在Burp项目中捕获的HTTP流量。
来源技能章节菜单:
- Burp Project Parser
- When to Use
- Prerequisites
- Quick Reference
- Resolving Exit 3: the control query
- Sub-Component Filters (USE THESE)
- Available Filters
- Default Approach
- Regex Search Operations
- Search Response Headers
- Search Response Bodies
- Other Operations
- Extract Audit Items
- Dump Proxy History (AVOID)
- Dump Site Map (AVOID)
- Output Limits (REQUIRED)
trailofbits/claude-in-chrome-troubleshooting
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/README.md
功能简介: 诊断并修复 Claude 在 Chrome MCP 扩展程序连接问题
适用时机: 修复 mcp__claude-in-chrome__* 工具失败时出现的“浏览器扩展程序未连接”错误。在 Claude Code CLI 和 Claude.app 之间切换活跃的原生消息主机。Cowork。在更新后将 chrome-native-host 包装程序重置为动态选择最新的 Claude Code 版本。诊断正在运行的哪个原生主机二进制文件,以及它是否在预期位置创建了套件。解决运行多个 Claude Code 会话或 Chrome 配置文件所致的套件冲突。
来源技能章节菜单:
- Trail of Bits Skills Marketplace
- Installation
- Claude Code Marketplace
- Browse and Install Plugins
- Codex
- ChatGPT Workspace Marketplace
- Local Development
- Available Plugins
- Smart Contract Security
- Code Auditing
- Malware Analysis
- Verification
- Reverse Engineering
- Mobile Security
- Development
- Team Management
trailofbits/constant-time-analysis
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/constant-time-analysis/skills/constant-time-analysis/SKILL.md
功能简介: 检测加密代码中编译器引起的时序侧向通道
适用时机: 在实施或审查加密代码时使用,遇到在C、C++、Go、Rust、Swift、Java、Kotlin、C#、PHP、JavaScript、TypeScript、Python或Ruby中在秘密上进行除法、依赖秘密的分支或常数时间编程问题时。
来源技能章节菜单:
- Constant-Time Analysis
- When to Use
- When NOT to Use
- Language Routing
- Running the Analyzer
- Per-language coverage limits
- Prerequisites
- Interpreting Results
- Triaging Findings
- Limitations
- Real-World Impact
- References
trailofbits/differential-review
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/differential-review/skills/differential-review/SKILL.md
功能简介: 以安全为重点的差异化审查,结合git历史记录分析
适用时机: 用于审查PR、提交或差异图以查找安全漏洞,检查更改是否重新引入了之前修复的错误,询问更改还可能破坏哪些功能,或查找哪些修改的代码没有覆盖测试。
来源技能章节菜单:
- Differential Security Review
- Core Principles
- Rationalizations (Do Not Skip)
- Quick Reference
- Codebase Size Strategy
- Risk Level Triggers
- Workflow Overview
- Decision Tree
- Agents
- Quality Checklist
- Integration
- Example Usage
- Quick Triage (Small PR)
- Standard Review (Medium Codebase)
- Deep Audit (Large, Critical Change)
- When NOT to Use This Skill
trailofbits/dwarf-expert
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/dwarf-expert/skills/dwarf-expert/SKILL.md
功能简介: DWARF调试格式专业知识
适用时机: 在使用dwarfdump/llvm-dwarfdump或readelf检查.debug_*部分、DIE树或DW_TAG_/DW_AT_条目时使用,使用llvm-dwarfdump --verify验证调试信息,回答DWARF标准问题,或编写解析DWARF的代码(libdwarf、pyelftools、gimli)。
来源技能章节菜单:
- DWARF Expert
- Authoritative Sources
- Parsing and Searching with dwarfdump
- Searching DIEs
- Verifying DWARF Integrity
- readelf
- Writing Code That Parses DWARF
trailofbits/entry-point-analyzer
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/entry-point-analyzer/skills/entry-point-analyzer/SKILL.md
功能简介: 识别智能合约中的状态变更入口点
适用时机: 在审计智能合约(Solidity、Vyper、Solana/Rust、Move、TON、CosmWasm)时使用,或在被要求查找入口点、审计流程、外部函数、访问控制模式或特权操作时使用。
来源技能章节菜单:
- Entry Point Analyzer
- Execution in Codex
- When to Use
- When NOT to Use
- Scope: State-Changing Functions Only
- Workflow
- Slither Integration (Solidity)
- 1. Check if Slither is Available
- 2. If Slither is Detected, Run Entry Points Printer
- 3. Use Slither Output as Foundation
- 4. When Slither is NOT Available
- Language Detection
- Access Classification
- 1. Public (Unrestricted)
- 2. Role-Restricted
- 3. Contract-Only (Internal Integration Points)
trailofbits/firebase-apk-scanner
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/firebase-apk-scanner/skills/firebase-apk-scanner/SKILL.md
功能简介: 扫描Android APK以查找Firebase的配置错误和安全漏洞
适用时机: 用于分析 Firebase 安全漏洞的 APK 文件、进行移动应用程序安全审计或测试 Firebase 端点安全时使用。仅供授权的安全研究使用。
来源技能章节菜单:
- Firebase APK Security Scanner
- When to Use
- When NOT to Use
- Rationalizations to Reject
- Reference Documentation
- How to Use This Skill
- Workflow
- Step 1: Validate Input
- Step 2: Run the Scanner
- Step 3: Present Results
- Scan Summary
- Extracted Configuration
- Vulnerabilities Found
- Remediation
- Manual Testing (If Scanner Fails)
- Severity Classification
trailofbits/insecure-defaults
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/insecure-defaults/README.md
功能简介: 检测不安全的默认配置,例如硬编码的秘密信息、默认凭证和弱密码学
适用时机: 审计 JWT 密钥处理,以捕获生产认证代码中的重置值,如 `|| ‘default’`。扫描 Dockerfile 和 IaC 模板,查找触发不安全运行时行为的缺失环境变量。在部署前审查环境变量处理,将 fail-open 与 fail-secure 模式分离。检查密码哈希和令牌签名上下文中的加密使用情况,检查是否存在 MD5、SHA1 或 ECB。追踪 CORS 乱序字符和权限 0777 的默认设置,以确认它们是否到达生产端点。
来源技能章节菜单:
- Insecure Defaults Detection
- Install
- Use
- What it finds
- How it runs
- Patterns
- When not to use it
- Tests
trailofbits/modern-python
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/modern-python/skills/modern-python/SKILL.md
功能简介: 使用uv、ruff、ty和pytest的最佳实践的现代Python工具
适用时机: 在创建项目、编写独立脚本或从pip/Poetry/mypy/black迁移时使用。
来源技能章节菜单:
- Modern Python
- When to Use This Skill
- When NOT to Use This Skill
- Anti-Patterns to Avoid
- Decision Tree
- Tool Overview
- Security Tools
- Quick Start: Minimal Project
- Full Project Setup
- 1. Create Project Structure
- 2. Configure pyproject.toml
- 3. Install Dependencies
- 4. Add Makefile
- Migration Guide
- From requirements.txt + pip
- From setup.py / setup.cfg
trailofbits/property-based-testing
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/property-based-testing/skills/property-based-testing/SKILL.md
功能简介: 针对多个语言和智能合约的基于属性的测试
适用时机: 适用于将案例添加到现有 @given、快速验证或 proptest 测试套件时,用于判断现有属性测试是否断言了任何真实内容,以及当生成器缩小了反例范围,而你需要区分错误属性与真实错误时。不适用于覆盖度引导的二进制伪随机测试(libFuzzer、AFL)、突变测试活动、静态分析、基准测试或端到端 UI 测试。
来源技能章节菜单:
- Property-Based Testing
- Property catalog
- The two ways a property test asserts nothing
- Where to look next
- Introducing PBT to a project that lacks it
trailofbits/semgrep-rule-creator
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/semgrep-rule-creator/skills/semgrep-rule-creator/SKILL.md
功能简介: 创建和完善用于漏洞检测的Semgrep规则
适用时机: 在编写Semgrep规则或构建自定义静态分析检测时使用。
来源技能章节菜单:
- Semgrep Rule Creator
- When to Use
- When NOT to Use
- Rationalizations to Reject
- Anti-Patterns
- Strictness Level
- Overview
- Quick Start
- Quick Reference
- Workflow
- Documentation
trailofbits/semgrep-rule-variant-creator
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/semgrep-rule-variant-creator/skills/semgrep-rule-variant-creator/SKILL.md
功能简介: 使用测试驱动验证将现有的Semgrep规则移植到新的目标语言中
适用时机: 用于将 Semgrep 规则移植到指定的目标语言。输入现有的规则和目标语言,为每个语言生成独立的规则+测试目录。
来源技能章节菜单:
- Semgrep Rule Variant Creator
- Run it as a workflow
- The four phases
- Output
- Scope and reporting
- Rationalizations to Reject
- Quick Reference
- Documentation
trailofbits/sharp-edges
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/sharp-edges/skills/sharp-edges/SKILL.md
功能简介: 识别容易出错的API和危险的配置
适用时机: 用于审查 API 设计、配置规范、加密库人机界面设计,或评估代码是否遵循“默认安全”和“成功陷阱”原则。触发因素:枪炮式设计、防滥用、安全默认设置、API 可用性、危险配置。
来源技能章节菜单:
- Sharp Edges Analysis
- When to Use
- When NOT to Use
- Agent
- Core Principle
- Rationalizations to Reject
- Sharp Edge Categories
- 1. Algorithm/Mode Selection Footguns
- 2. Dangerous Defaults
- 3. Primitive vs. Semantic APIs
- 4. Configuration Cliffs
- 5. Silent Failures
- 6. Stringly-Typed Security
- Analysis Workflow
- Phase 1: Surface Identification
- Phase 2: Edge Case Probing
trailofbits/spec-to-code-compliance
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/spec-to-code-compliance/skills/spec-to-code-compliance/SKILL.md
功能简介: 用于区块链审计的规范到代码合规检查器
适用时机: 用于比较一个实施方案与白皮书、协议规范或设计文档。
来源技能章节菜单:
- Spec-to-Code Compliance
- When to Use
- When NOT to Use
- Do not check requirements in this context
- What comes back, and how to read it
- Judgment the workflow does not make for you
- The target does not have to be a contract
- Reference
trailofbits/static-analysis
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/static-analysis/README.md
功能简介: 包含CodeQL、Semgrep和SARIF的静态分析工具包
适用时机: 运行 CodeQL 污染分析以追踪用户输入到 SQL 查询中 编写 Semgrep 规则以在整个代码库中捕获特定 CWE 模式 将多个扫描器的 SARIF 输出解析为去重发现列表 使用 Semgrep 对 Pull Request 进行 OWASP 和 Trail of Bits 规则集扫描 为项目特定的 API 汇流口生成 CodeQL 数据扩展模型
来源技能章节菜单:
- Static Analysis
- Skills Included
- When to Use
- What It Does
- CodeQL
- Semgrep
- SARIF Parsing
- Running a Semgrep scan
- Workflows Included
- Scripts Included
- Tests
- Installation
- Related Skills
此目录条目指向一个技能集合。扫描来源仓库发现 3 个不重复的技能文件路径。
部分成员: codeql · sarif-parsing · semgrep
trailofbits/testing-handbook-skills
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/testing-handbook-skills/README.md
功能简介: 测试手册技能:伪装器、静态分析、净化器
适用时机: 直接从手册的 fuzzing 章节中生成 libFuzzer 技能 使用手册准确的规则编写指导构建 Semgrep 技能的框架 编写一个跨引用 sanitizer 和覆盖率技能的 harness 编写技能 在 CI 中运行 validate-skills.py 以捕获生成技能之间的断开跨引用 从一个命令中构建一套完整的 16 个协调的安全测试技能
来源技能章节菜单:
- Testing Handbook Skills
- Overview
- Installation
- Usage
- Generate All Skills
- Generate Specific Skill
- Structure
- Scripts
- Skill Types
- Generated Skills
- Skills Cross-Reference
- Configuration
- Author
- License
此目录条目指向一个技能集合。扫描来源仓库发现 15 个不重复的技能文件路径。
部分成员: address-sanitizer · aflpp · atheris · cargo-fuzz · constant-time-testing · coverage-analysis · fuzzing-dictionary · fuzzing-obstacles
更多成员请查看来源仓库。
trailofbits/variant-analysis
目录来源: Security Skills by Trail of Bits Team
来源文档: https://github.com/trailofbits/skills/blob/main/plugins/variant-analysis/skills/variant-analysis/SKILL.md
功能简介: 通过模式分析找到类似的漏洞
适用时机: 在在认证处理器中找到一个后,从整个代码库中搜索null等价性绕过机制 从已知XSS模式构建Semgrep规则,并迭代直到覆盖相关泄露点 将单个根本原因中的30个ripgrep匹配结果分类为高/中/低置信度发现结果 从Python注入漏洞扩展CodeQL查询,覆盖同一仓库中的Java和Go变体 在CVE补丁后执行五步变体猎捕,以在分叉的代码路径中发现遗漏的实例
来源技能章节菜单:
- Variant Analysis
- When to Use
- When NOT to Use
- The Five Steps
- Running it as a Workflow
- What Makes Hunts Fail
- Resources
veniceai/venice-api-overview
目录来源: Skills by Venice.ai
来源文档: https://github.com/veniceai/skills/blob/main/skills/venice-api-overview/SKILL.md
功能简介: API基础知识、认证模式、定价和版本化
适用时机: 当任务需要此目录条目中描述的能力时使用:API基础知识、认证模式、定价和版本化
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Venice API Overview
- Use when
- Base URL
- Authentication
- Endpoint map
- Inference
- Catalog
- Account, billing, wallet
- Utility
- Retired (return `410 Gone`)
- Response headers to watch
- Pricing model at a glance
- Standard error shape
- OpenAI compatibility - what works and what doesn’t
- Versioning
- Fast start checklist
veniceai/venice-auth
目录来源: Skills by Venice.ai
来源文档: https://github.com/veniceai/skills/blob/main/skills/venice-auth/SKILL.md
功能简介: API密钥和基于钱包的威尼斯身份验证
适用时机: 当任务需要此目录条目中描述的能力时使用:API密钥和基于钱包的威尼斯身份验证
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Venice Authentication
- Use when
- Which endpoints accept what
- Option A - Bearer API key
- Option B - x402 wallet (SIWX)
- Header
- SIWE message fields (EIP-4361, EVM)
- Solana message fields
- Manual signing (TypeScript, EVM)
- SDK shortcut (EVM / Base)
- First-time top-up (wallet → credits)
- Choosing between the two
- Common auth errors
- Security hygiene
- Payment safety
vercel-labs/vercel-cli-with-tokens
目录来源: Skills by Vercel Engineering Team
来源文档: https://github.com/vercel-labs/agent-skills/blob/main/skills/vercel-cli-with-tokens/SKILL.md
功能简介: 使用代币身份验证部署和管理Vercel项目
适用时机: 在使用Vercel CLI时使用访问令牌而不是交互式登录——例如“部署到vercel”、“设置vercel”、“将环境变量添加到vercel”。
来源技能章节菜单:
- Vercel CLI with Tokens
- Step 1: Locate the Vercel Token
- A) `VERCEL_TOKEN` is already set in the environment
- B) Token is in a `.env` file under `VERCEL_TOKEN`
- C) Token is in a `.env` file under a different name
- D) No token found — ask the user
- Step 2: Locate the Project and Team
- CLI Setup
- Deploying a Project
- Quick Deploy (have project ID — no linking needed)
- Full Deploy Flow (no project ID — need to link)
- Deploying from a Remote Repository (code not cloned locally)
- About `.vercel/` Directory
- Managing Environment Variables
- Inspecting Deployments
- Managing Domains
VoDaiLocz/kilo-kit-mcp
目录来源: Development and Testing
来源文档: https://github.com/VoDaiLocz/kilo-kit-mcp/blob/main/README.md
功能简介: 包含177项精心挑选技能的综合图书馆,搭配一个强制执行协议级C4工作流程闸道的MCP运行时环境,配备安全防护栏杆的硬闸命令执行功能,以及5个认知推理引擎(Tree of Thoughts DAG、Adversarial Grilling、5-Whys根因追踪器、上下文压缩器、自我进化),适用于Claude Code、Cursor、Antigravity和Codex。
适用时机: 当任务需要本目录条目中描述的能力时使用:包含177项精心挑选技能的综合库,搭配一个强制执行协议级C4工作流程闸道的MCP运行时环境,配备安全防护栏杆的硬闸命令执行功能,以及5个认知推理引擎(Tree of Thoughts DAG、Adversarial Grilling、5-Whys根因追踪器、Context Compactor、Self-Evolution),适用于Claude Code、Cursor、Antigravity和Codex。
来源目录没有单独说明触发条件;此处的适用时机是根据公开功能简介归纳的。
来源技能章节菜单:
- Kilo-Kit
- 💡 Why Kilo-Kit?
- The Crisis of Unbounded AI Agency
- The Solution: Governed Autonomy
- ⚡ Quick Start (in 30 seconds)
- 1. Zero-Install Instant Setup
- 2. Verify System Health
- 🏛️ Architecture & C4 Protocol
- The C4 5-Gate Lifecycle
- 💻 IDE Integration Guide
- ⚡ Core Capabilities
- 1. 🚦 4-Tier Adaptive Task Gating
- 2. 🔍 SQLite FTS5 with BM25 Lexical Ranking
- 3. ✍️ Self-Authoring Skills & Autonomous Evolution
- 4. 🛡️ Kilo-Sentinel Loop & Thrashing Tripwire
- 🧰 MCP Tools Suite (24 Tools)
此目录条目指向一个技能集合。扫描来源仓库发现 168 个不重复的技能文件路径。
部分成员: agent-memory · claukit · mcp-agent-patterns · multi-agent-orchestration · workflow-state-machines · ai-multimodal · geo-fundamentals · media-processing
更多成员请查看来源仓库。
WordPress/wp-abilities-api
目录来源: Skills by WordPress Development Team
来源文档: https://github.com/WordPress/agent-skills/blob/trunk/skills/wp-abilities-api/SKILL.md
功能简介: 基于能力的权限和REST API认证
适用时机: 在与WordPress能力API(wp_register_ability、wp_register_ability_category、/wp-json/wp-abilities/v1/*、@wordpress/abilities)一起工作时使用,包括定义能力、类别、元数据、REST暴露和客户端权限检查。
来源技能章节菜单:
- WP Abilities API
- When to use
- Inputs required
- Procedure
- 1) Confirm availability and version constraints
- 2) Find existing Abilities usage
- 3) Register categories (optional)
- 4) Register abilities (PHP)
- 5) Confirm REST exposure
- 6) Consume from JS (if needed)
- Verification
- Failure modes / debugging
- Escalation
WordPress/wp-plugin-development
目录来源: Skills by WordPress Development Team
来源文档: https://github.com/WordPress/agent-skills/blob/trunk/skills/wp-plugin-development/SKILL.md
功能简介: 插件架构、钩子、设置API、安全
适用时机: 在开发WordPress插件时使用:架构和钩子,激活/停用/卸载,管理员UI和设置API,数据存储,cron/任务,安全(非重复标识符/能力/消毒/逃脱),以及发布包装。
来源技能章节菜单:
- WP Plugin Development
- When to use
- Inputs required
- Procedure
- 0) Triage and locate plugin entrypoints
- 1) Follow a predictable architecture
- 2) Hooks and lifecycle (activation/deactivation/uninstall)
- 3) Settings and admin UI (Settings API)
- 4) Security baseline (always)
- 5) Data storage, cron, migrations (if needed)
- Verification
- Failure modes / debugging
- Escalation
WordPress/wp-rest-api
目录来源: Skills by WordPress Development Team
来源文档: https://github.com/WordPress/agent-skills/blob/trunk/skills/wp-rest-api/SKILL.md
功能简介: REST API路由/端点、结构体、身份验证和响应整形
适用时机: 在构建、扩展或调试WordPress REST API端点/路由时使用:register_rest_route、WP_REST_Controller/控制器类、规范/参数验证、权限_回调/身份验证、响应整形、register_rest_field/register_meta,或通过show_in_rest来暴露CPTs/分类。
来源技能章节菜单:
- WP REST API
- When to use
- Inputs required
- Procedure
- 0) Triage and locate REST usage
- 1) Choose the right approach
- 2) Register routes safely (namespaces, methods, permissions)
- 3) Validate/sanitize request args
- 4) Responses, fields, and links
- 5) Authentication and authorization
- 6) Client-facing behavior (discovery, pagination, embeds)
- Verification
- Failure modes / debugging
- Escalation
wrsmith108/varlock-claude-skill
目录来源: Development and Testing
来源文档: https://github.com/wrsmith108/varlock-claude-skill/blob/main/skills/varlock/SKILL.md
功能简介: 安全的环境变量管理,确保秘密信息在 Claude 会话、终端、日志或 git 提交中永远不会泄露。
适用时机: 用于处理秘密信息、API密钥、凭证或任何敏感配置时使用。确保秘密信息永远不会在终端、日志、跟踪记录或Claude的上下文中泄露。触发短语包括“环境变量”、“秘密信息”、“.env”、“API密钥”、“凭证”、“敏感”、“Varlock”。
来源技能章节菜单:
- Varlock Security Skill
- Core Principle: Secrets Never Exposed
- CRITICAL: Security Rules for Claude
- Rule 1: Never Echo Secrets
- Rule 2: Never Read .env Directly
- Rule 3: Use Varlock for Validation
- Rule 4: Never Include Secrets in Commands
- Quick Start
- Installation
- Initialize Project
- Schema File: .env.schema
- Basic Structure
- Security Annotations
- Type Annotations
- Safe Commands for Claude
- Validating Environment
安全与隐私 | Agent Skills

