awesome-claude-skills — Claude Skill 中文使用文档
AI Skill Hub 强烈推荐：awesome-claude-skills — Claude Skill 中文使用文档 是一款优质的Claude技能。AI 综合评分 8.0 分，在同类工具中表现稳健。如果你正在寻找可靠的Claude技能解决方案，这是一个值得深入了解的选择。
**为什么要使用开源工具而非商业 SaaS？**
对于个人开发者和有隐私需求的用户，本地部署的开源工具意味着数据不离本机，不受第三方服务商的数据政策约束。同时，开源工具通常没有使用次数限制和月度费用，一次安装即可长期使用，对于高频使用场景的总拥有成本（TCO）远低于订阅制商业工具。
**安装与环境准备**
awesome-claude-skills — Claude Skill 中文使用文档 依赖 Python 运行环境。建议通过 pyenv（Python）或 nvm（Node.js）管理 Python 版本，避免全局环境污染。对于新手用户，推荐先创建虚拟环境（python -m venv venv && source venv/bin/activate），再安装依赖，这样即使出现问题也可以随时删除虚拟环境重新开始，不影响系统稳定性。
**社区与维护**
GitHub Issue 和 Discussion 是获取帮助的最快渠道。在提问前建议先检查 Closed Issues（已关闭的问题），大多数常见问题都已有解答。遇到 Bug 时，提供 pip list 的输出、完整错误堆栈和最小可复现示例，能显著提高开发者响应速度。AI Skill Hub 将持续追踪 awesome-claude-skills — Claude Skill 中文使用文档 的版本更新，及时通知重要功能变化。
awesome-claude-skills — Claude Skill 中文使用文档 是一款基于 Python 开发的开源工具，专注于 agent-skills、ai、anthropic 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
awesome-claude-skills — Claude Skill 中文使用文档 是一款基于 Python 开发的开源工具，专注于 agent-skills、ai、anthropic 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
- 专为 Claude 系列模型深度优化的扩展技能
- 通过 Claude Code CLI 一键安装，配置零门槛
- 充分利用 Claude 的长上下文和推理能力
- 支持与 MCP 工具组合使用，扩展能力边界
- 在 Claude 中快速解决特定专业领域的问题
- 复杂任务的 AI 辅助分析、推理和报告生成
- 构建个人专属的 AI 技能工具箱
# 克隆仓库 git clone https://github.com/karanb192/awesome-claude-skills cd awesome-claude-skills # 查看安装说明 cat README.md # 按 README 完成环境依赖安装后即可使用
- 访问 GitHub 仓库页面
- 按照 README 文档完成依赖安装
- 根据系统环境完成初始化配置
- 参考官方示例或文档开始使用
- 遇到问题可在 GitHub Issues 中查找解答
# 查看帮助 awesome-claude-skills --help # 基本运行 awesome-claude-skills [options] <input> # 详细使用说明请查阅文档 # https://github.com/karanb192/awesome-claude-skills
# awesome-claude-skills 配置说明 # 查看配置选项 awesome-claude-skills --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量（覆盖配置文件） export AWESOME_CLAUDE_SKILLS_CONFIG="/path/to/config.yml"
awesome-claude-skills
The definitive collection of Agent Skills for Claude - supercharge your AI workflows across Claude Code, Claude.ai, and API
Claude just got Skills! This is the definitive collection of 50+ Agent Skills for Claude Code, Claude.ai, and Claude API to boost productivity, enforce best practices, and automate complex workflows.
🎯 Why this list? Verified skills ✓ | Active maintenance ✓ | Clear use cases ✓ | Community-driven ✓ | 50+ Skills ✓
💡 New to Skills? Start with the Quick Start Guide • Looking for something specific? Use Ctrl+F to search • Want to contribute? Check Contributing
Featured Skills
Start with these top 10 essential skills:
| Skill | Why You Need It | Category | Verified |
|---|---|---|---|
| [test-driven-development](https://github.com/obra/superpowers) | Write bulletproof code with RED-GREEN-REFACTOR workflow | 🧪 Testing | ✅ |
| [systematic-debugging](https://github.com/obra/superpowers) | Find bugs 10x faster with 4-phase root cause analysis | 🐛 Debugging | ✅ |
| [using-git-worktrees](https://github.com/obra/superpowers) | Work on multiple features simultaneously without context switching | 🤝 Workflow | ✅ |
| [mcp-builder](https://github.com/anthropics/skills) | Build custom MCP servers to extend Claude's capabilities | ⚙️ Development | ✅ |
| [pdf](https://github.com/anthropics/skills) | Extract text, tables, metadata from PDFs with merge & annotation support | 📄 Documents | ✅ |
| [docx](https://github.com/anthropics/skills) | Create, edit, and analyze Word documents with tracked changes | 📄 Documents | ✅ |
| [artifacts-builder](https://github.com/anthropics/skills) | Build complex React artifacts with Tailwind CSS and shadcn/ui | ⚙️ Development | ✅ |
| [skill-creator](https://github.com/anthropics/skills) | Create your own skills and contribute to the ecosystem | 🎯 Meta | ✅ |
| [requesting-code-review](https://github.com/obra/superpowers) | Pre-review preparation with formatted diffs and clear PR descriptions | 🤝 Workflow | ✅ |
| [subagent-driven-development](https://github.com/obra/superpowers) | Quality-gated iteration with multi-agent workflows for complex tasks | 🎯 Meta | ✅ |
What's the difference between skills and MCP servers?
See the comparison table above. TL;DR: Skills for workflows, MCP for external tools.
1. Install obra's superpowers collection (20+ battle-tested skills)
git clone https://github.com/obra/superpowers ~/.claude/skills/superpowers
In Claude Code, just say: "Let's use TDD to build a user authentication system"
How to Install Skills
Method 2: Manual Installation
- Create a folder in
~/.claude/skills/ - Add a
SKILL.mdfile with YAML frontmatter and instructions - (Optional) Include supporting scripts and resources
Verify installation: ```bash
Quick Start
Get your first skill running in 30 seconds:
```bash
Claude will automatically load the TDD skill and guide you through RED-GREEN-REFACTOR!
```
That's it! Skills load automatically when relevant. No configuration needed.
🤝 Collaboration & Workflow
#### requesting-code-review Source: obra/superpowers | Verified: ✅ Description: Pre-review preparation and PR best practices with formatted diffs. Use Case: Before submitting PRs, preparing for team review Stars: ⭐⭐⭐⭐⭐
#### receiving-code-review Source: obra/superpowers | Verified: ✅ Description: Constructive feedback integration and iteration on review comments. Use Case: Responding to PR feedback, implementing requested changes Stars: ⭐⭐⭐⭐
#### using-git-worktrees Source: obra/superpowers | Verified: ✅ Description: Parallel development branches for context switching optimization. Use Case: Juggling multiple features, emergency hotfixes, experimental branches Stars: ⭐⭐⭐⭐⭐
#### finishing-a-development-branch Source: obra/superpowers | Verified: ✅ Description: Guides merge/PR decisions and maintaining clean git history. Use Case: Preparing features for merge, cleaning up commit history Stars: ⭐⭐⭐⭐
#### brainstorming Source: obra/superpowers | Verified: ✅ Description: Socratic design refinement and feature exploration through guided questioning. Use Case: Architecture decisions, API design, feature planning Stars: ⭐⭐⭐⭐
#### writing-plans Source: obra/superpowers | Verified: ✅ Description: Creates detailed implementation strategies and architecture documentation. Use Case: Complex features, system design, technical specs Stars: ⭐⭐⭐⭐⭐
#### executing-plans Source: obra/superpowers | Verified: ✅ Description: Batch execution with checkpoints for progress tracking and recovery. Use Case: Large refactors, multi-step implementations Stars: ⭐⭐⭐⭐
---
Skills vs MCP vs System Prompts
Confused about when to use Skills vs other Claude customization methods? Here's the breakdown:
| Feature | Skills | MCP Servers | System Prompts |
|---|---|---|---|
| **Purpose** | Task-specific workflows | External tool integration | General behavior modification |
| **Setup** | Git clone to ~/.claude/skills/ | Install & configure MCP server | Edit CLAUDE.md in project |
| **Activation** | Automatic (context-aware) | Explicit tool calls | Always active |
| **Best For** | TDD, debugging, git workflows | APIs, databases, file systems | Project conventions, style guides |
| **Portability** | Cross-platform (CLI, web, API) | Platform-dependent | Project-specific |
| **Token Cost** | 30-50 until loaded | Per-call | Always consuming tokens |
| **Examples** | test-driven-development | Weather API, GitHub integration | "Use TypeScript strict mode" |
When to use what: - ✅ Skills → Repeatable workflows (TDD, debugging, code review) - ✅ MCP → External data/tools (APIs, search, databases) - ✅ System Prompts → Project-specific rules and conventions
🐛 Debugging & Troubleshooting
#### systematic-debugging Source: obra/superpowers | Verified: ✅ Description: Four-phase root cause process: reproduce, isolate, identify, verify fix. Use Case: Complex bugs, production issues, multi-component failures Stars: ⭐⭐⭐⭐⭐
#### root-cause-tracing Source: obra/superpowers | Verified: ✅ Description: Deep problem investigation with dependency chain analysis. Use Case: Tracing cascading failures, understanding system interactions Stars: ⭐⭐⭐⭐
#### verification-before-completion Source: obra/superpowers | Verified: ✅ Description: Ensures fixes are validated before marking work complete. Use Case: Bug fixes, refactoring work, feature additions Stars: ⭐⭐⭐⭐
#### defense-in-depth Source: obra/superpowers | Verified: ✅ Description: Multiple validation layers for comprehensive error handling. Use Case: Critical systems, production code, API endpoints Stars: ⭐⭐⭐
#### performance-profiling Status: Community-needed Description: Identify performance bottlenecks, memory leaks, and CPU-intensive operations. Use Case: Optimization work, scaling applications, investigating slowness
---
FAQ
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- CLI：直接 npm install -g / pip install，命令行调用
- 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
- 专为 Claude 系列模型深度优化的扩展技能
- 通过 Claude Code CLI 一键安装，配置零门槛
- 充分利用 Claude 的长上下文和推理能力
- 支持与 MCP 工具组合使用，扩展能力边界
- 在 Claude 中快速解决特定专业领域的问题
- 复杂任务的 AI 辅助分析、推理和报告生成
- 构建个人专属的 AI 技能工具箱
- +MIT 协议，可免费商用
- +深度优化 Claude 使用体验
- +CLI 一键安装，极度便捷
- +官方支持，稳定可靠
- −仅限 Claude 用户使用，受平台限制
- −功能边界受当前 Claude 模型能力约束
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一，可自由商用、修改、分发，仅需保留版权声明。
总体来看，awesome-claude-skills — Claude Skill 中文使用文档 是一款质量优秀的Claude技能，在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态，建议收藏备用，结合自身场景选择合适时机引入使用。
| 原始名称 | awesome-claude-skills |
| 原始描述 | 🎯 The definitive collection of 50+ verified Awesome Claude Skills for Claude Code, Claude.ai, and API. Boost productivity with TDD, debugging, git workflows, document processing, and more. Community-driven, actively maintained. |
| Topics | agent-skillsaianthropicanthropic-claudeartifical-intelligenceautomationoffice |
| GitHub | https://github.com/karanb192/awesome-claude-skills |
| License | MIT |
收录时间：2026-05-22 · 更新时间：2026-05-22 · License：MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。