Skill Title【免费下载链接】agentic-awesome-skillsAAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,115 agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.项目地址: https://gitcode.com/gh_mirrors/an/agentic-awesome-skills标题应清晰、有描述性通常与技能名一致或在其基础上扩展。 ### 2. Overview概述 24 句话说明这个技能做什么、为什么存在。模板要求如果技能改编自外部 GitHub 仓库在此处同时声明 source_repo 与 source_type原创技能则使用 source: self 与 source_type: self。 ### 3. When to Use This Skill触发时机 这是**质量门槛强制要求**的章节——机器并不检查正文里的其他章节但必须有明确触发时机说明是硬性门槛见 [docs/contributors/quality-bar.md](https://link.gitcode.com/i/c988f867ecdeed0d2b3e0005090033a1) 第 2 条。验收的标题包括 ## When to Use、## Use this skill when、## When to Use This Skill 三种写法。 markdown ## When to Use This Skill - Use when you need to [scenario 1] - Use when working with [scenario 2] - Use when the user asks about [scenario 3]好的写法当用户要求调试 React 组件时使用差的写法这个技能帮你处理代码。4. How It Works核心指令## How It Works ### Step 1: [Action] Detailed instructions... ### Step 2: [Action] More instructions...这是技能的心脏必须是清晰、可执行的分步指令。技能解剖文档 docs/contributors/skill-anatomy.md 给出三条写作铁律用直接语言不要写 You might want to consider possibly checking if the user has authentication直接写 Check if the user is authenticated before proceeding.用祈使动词写 Create the file...而不是 The file should be created...具体到命令不要写 Set up the database properly而是列出一、二、三步建库、npm run migrate、npm run seed。5. Examples示例## Examples ### Example 1: [Use Case] javascript // Example code质量门槛要求至少包含一个**可复制粘贴**的代码块或交互示例用户或 Agent 拿到即可用。示例向 Agent 展示好的输出长什么样。 ### 6. Best Practices最佳实践 用 ✅ / ❌ 清单呈现做与不做 markdown ## Best Practices - ✅ Do this - ✅ Also do this - ❌ Dont do this - ❌ Avoid this7. Limitations限制列出已知边界与技能不能做的事。质量门槛将其列为独立质量项A list of known edge cases or things the skillcannotdo示例Does not work on Windows without WSL.。模板还额外强调当必需的输入、权限或安全边界缺失时应停下来向用户澄清而不是强行执行。8. Security Safety Notes安全与操作注记凡技能包含 shell 命令或命令式示例、网络拉取、token/能力字符串、或直接的变更引导都必须显式写出前置条件与警示。模板给出了一个关键的允许清单注释机制!-- security-allowlist: approved for documented workflow X --对于有意保留的高风险示例如curl ... | bash、wget ... | sh、凭据示例需要给出评审可见的理由并加上上述 allowlist 注释。质量门槛文档说明npm run security:docs实现于 tools/scripts/tests/docs_security_content.test.js会在仓库范围内扫描三类内容curl ... | bash类管道、内联 token/密钥式命令示例、以及通过 allowlist 注释放行的刻意高风险命令。能够修改文件/系统或执行危险操作的技能还要写明确认门confirmation gate与环境预期如local-only、authorized test environment。9. Common Pitfalls常见坑## Common Pitfalls - **Problem:** Description **Solution:** How to fix it10. Related Skills相关技能## Related Skills - other-skill - When to use this instead - complementary-skill - How this works together这组引用让 Agent 能在相关技能之间做交叉调度。例如技能解剖文档给出的工作流编排示例先用brainstorming做设计再用writing-plans制定计划最后用test-driven-development实现。进攻性技能的硬性红线如果新技能属于risk: offensive类别除了模板中的常规章节外还受 docs/contributors/security-guardrails.md 的三重约束必须原样包含授权声明且校验器会按正则强制匹配见 validate_skills.py **⚠️ AUTHORIZED USE ONLY** This skill is for educational purposes or authorized security assessments only. You must have explicit, written permission from the system owner before using this tool. Misuse of this tool is illegal and strictly prohibited.禁止全自动执行每次针对目标的探测、利用、变更、持久化、数据提取或凭据访问命令前Agent 必须收集确切目标、书面授权确认与允许范围展示确切命令与预期影响并在当前对话中等待用户明确确认——校验器同样用正则检查 Mandatory confirmation gate、exact target URL, IP, account, or resource、Wait for explicit confirmation in the current conversation 等关键句见 validate_skills.py。未获确认前技能必须保持只读并仅提供防御性指导。安全设计不包含可用作武器的载荷恶意软件、勒索软件、非教育用途的漏洞利用并建议在隔离环境Docker/VM中运行。即使是防御性技能safe/critical也不得在未经用户明确同意的情况下向第三方服务器上传数据审计类操作默认只读。配套资源目录的使用规范当技能需要脚本、示例、模板或参考文档时按固定目录组织并在 SKILL.md 中引用scripts/辅助脚本如setup.sh、validate.py、generate.js在正文中给出运行命令例如bash scripts/setup.shexamples/真实可用示例basic-usage.js、advanced-pattern.ts支持带config.json的完整实现子目录templates/可复用代码模板可用{{#include templates/component.tsx}}语法内联到 SKILL.mdreferences/外部文档或 API 参考api-docs.md、troubleshooting.md。篇幅分级从最小可行到全面型模板虽未直接分级但技能解剖文档给出了三种规模参考便于按内容复杂度取舍级别Frontmatter内容篇幅必备章节最小可行技能name/description/category/risk/source/date_added100–200 词Overview Instructions标准技能同上300–800 词Overview When to Use Instructions Examples全面型技能标准字段 source_repo/source_type及可选字段800–2000 词全部推荐章节可带 scripts/examples/templates经验法则从小开始依据反馈迭代。npm run fix:missing-sections对应 tools/scripts/fix_missing_skill_sections.py这类修复脚本可以辅助补齐缺失章节但人工校验仍是最终保障。写作技巧与触发优化先写 When to Use它最能厘清技能定位先写示例再写说明示例能帮你发现自己真正在教什么交给 AI 实测提交前用你的 Agent 实际运行一次验证触发与执行效果用渐进披露## Basic Usage放常规用法## Advanced Usage放进阶模式用条件逻辑例如 If the user is working with React: ... / If the user is working with Vue: ...让一个技能在不同技术栈下给出差异化指令跨技能编排在 Related Skills 中用引用串联工作流。技能解剖文档还给出了一个真实案例分析skills/brainstorming/SKILL.md强描述性You MUST use this before any creative work...、清晰标题、分阶段步骤是值得模仿的范本。仓库内可参照的其他范例包括skills/git-pushing/SKILL.md简单聚焦、skills/copywriting/SKILL.md示例优秀、skills/systematic-debugging/SKILL.md综合全面、skills/loki-mode/SKILL.md复杂工作流。提交前的验证命令全流程模板文档指向的验证体系在 package.json 中暴露为 npm scripts。完整的提交前验证序列如下npm run validate npm run audit:skills npm run validate:references npm test npm run security:docs【免费下载链接】agentic-awesome-skillsAAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,115 agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.项目地址: https://gitcode.com/gh_mirrors/an/agentic-awesome-skills创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考