Skip to content
Back to search

awesome-skills

🌟1000+ Expert AI Skills | CEO, Doctor, Engineer, Scientist & more | Transform AI into any professional | Powered by https://theneoai.github.io/skill-writer/

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/theneoai--awesome-skills/badge.svg)](/listings/theneoai--awesome-skills)

HTML

<a href="/listings/theneoai--awesome-skills"><img src="/api/listings/theneoai--awesome-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

The simplest way to install a skill is to have your agent read its `SKILL.md` URL:

Read https://github.com/theneoai/awesome-skills/blob/main/skills/persona/executive/ceo/SKILL.md and install as a skill

Platform-specific instructions (OpenCode native command, Claude Code, Cursor, Cline, Codex, Kimi) are in **[INSTALL-GUIDE.md](./INSTALL-GUIDE.md)**.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 310 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <!-- Awesome Skills -->
  2. <!-- Language: [English](#english) | [中文](#中文) -->
  3. <div align="center">
  4. # Awesome Skills
  5. **A curated library of role-based AI skills, organised by professional domain.**
  6. [![Skills](https://img.shields.io/badge/skills-956-blueviolet?style=flat-square)](./CATALOG.md)
  7. [![License](https://img.shields.io/badge/license-MIT-green?style=flat-square)](./LICENSE)
  8. [![Quality](https://github.com/theneoai/awesome-skills/actions/workflows/quality.yml/badge.svg)](https://github.com/theneoai/awesome-skills/actions/workflows/quality.yml)
  9. [![Evaluation](https://github.com/theneoai/awesome-skills/actions/workflows/comprehensive-evaluation.yml/badge.svg)](https://github.com/theneoai/awesome-skills/actions/workflows/comprehensive-evaluation.yml)
  10. [![Pages](https://github.com/theneoai/awesome-skills/actions/workflows/pages-deploy.yml/badge.svg)](https://github.com/theneoai/awesome-skills/actions/workflows/pages-deploy.yml)
  11. **[🌐 Website](https://theneoai.github.io/awesome-skills/)** · **[📚 Catalog](./CATALOG.md)** · **[📖 Install Guide](./INSTALL-GUIDE.md)** · **[🤝 Contributing](./CONTRIBUTING.md)**
  12. **Language:** [English](#english) | [中文](#中文)
  13. </div>
  14. ---
  15. <a name="english"></a>
  16. ## English
  17. ### What this is
  18. A library of **956 skill files** (`SKILL.md` + optional `references/`) covering 3 kinds (persona / tool / workflow) across 60+ domains. Each skill is a prompt pack — an instruction set plus reference material — designed to load into Claude Code, OpenCode, Cursor, or similar agent runtimes to give an LLM a consistent professional persona, tool expertise, or engineering workflow methodology.
  19. ### What this is NOT
  20. - **Not an alternative to executable skill packs.** The official Anthropic/OpenAI `SKILL.md` convention can ship with `scripts/` and `assets/` that the agent runs. The vast majority of skills here are **persona + methodology**, not runnable tools. Treat them as opinionated system prompts, not as plugins.
  21. - **Not a certification body.** Historical quality scores in this repo were produced by a self-authored scoring script grading against a self-authored rubric. The scoring is useful internally; it is not an independent benchmark.
  22. - **Not feature-complete for every domain.** Coverage is broad but uneven — some categories have deep role hierarchies, others have a single stub.
  23. ### Layout
  24. ```
  25. skills/ 956 skill files organised by kind
  26. persona/ 827 role-based professional persona skills
  27. <domain>/<role>/
  28. SKILL.md Frontmatter + system prompt (target ≤ 300 lines)
  29. references/ On-demand deep content (workflow / scenarios / …)
  30. EVALUATION_REPORT.md Optional self-scored quality report
  31. tool/ 116 technology-specific expert skills (kind: tool)
  32. <technology>/
  33. <skill-name>/SKILL.md
  34. workflow/ 8 process-driven action skills (kind: workflow)
  35. engineering/ TDD, debug-diagnose, zoom-out, architecture-review,
  36. issue-triage, to-prd
  37. meta/ write-skill, caveman
  38. external/ Registry of curated third-party skill repos
  39. (anthropics/skills, mattpocock/skills, VoltAgent, …).
  40. Pulled on demand via scripts/sync_external.py — not vendored.
  41. benchmarks/ Evaluation dataset + scoring script for comparison
  42. packages/ Curated skill bundles by domain
  43. roadmap/ Career-path documents (independent of skills/)
  44. taxonomy.yml Single source of truth: 21 categories, 80 aliases
  45. tools/ Python package for skill analysis (scoring, tokens, anti-patterns)
  46. scripts/ Maintenance scripts (catalog regen, external sync, taxonomy check, …)
  47. .github/ CI workflows + scripts, CI/CD docs
  48. ```
  49. ### Quick start
  50. The simplest way to install a skill is to have your agent read its `SKILL.md` URL:
  51. ```
  52. Read https://github.com/theneoai/awesome-skills/blob/main/skills/persona/executive/ceo/SKILL.md and install as a skill
  53. ```
  54. Platform-specific instructions (OpenCode native command, Claude Code, Cursor, Cline, Codex, Kimi) are in **[INSTALL-GUIDE.md](./INSTALL-GUIDE.md)**.
  55. ### Skill packages
  56. Pre-bundled collections in `packages/`:
  57. | Package | Focus |
  58. |---------|-------|
  59. | [tech](./packages/tech.md) | Software, AI/ML, data |
  60. | [executive](./packages/executive.md) | CEO, CTO, CFO, COO, CMO |
  61. | [finance](./packages/finance.md) | Banking, consulting, investment |
  62. | [healthcare](./packages/healthcare.md) | Clinical and medical management |
  63. | [software](./packages/software.md) | Backend, frontend, devops, QA |
  64. ### Featured enterprise skills
  65. Role skills modelled after the methodology of specific companies (15 shown out of 100+ under `skills/enterprise/`):
  66. | Skill | Company | Methodology |
  67. |-------|---------|------------|
  68. | [amazon-engineer](./skills/persona/enterprise/amazon/amazon-engineer/SKILL.md) | Amazon | 14 LPs, Working Backwards, 6-page memos |
  69. | [tesla-engineer](./skills/persona/enterprise/tesla/tesla-engineer/SKILL.md) | Tesla | First principles, five-step algorithm |
  70. | [spacex-engineer](./skills/persona/enterprise/spacex/spacex-engineer/SKILL.md) | SpaceX | Rapid iteration, cost innovation |
  71. | [nvidia-ml-engineer](./skills/persona/enterprise/nvidia/nvidia-ml-engineer/SKILL.md) | NVIDIA | CUDA optimisation, GPU platforms |
  72. | [mckinsey-consultant](./skills/persona/enterprise/mckinsey/mckinsey-consultant/SKILL.md) | McKinsey | MECE, issue trees, pyramid principle |
  73. | [toyota-engineer](./skills/persona/enterprise/toyota/toyota-engineer/SKILL.md) | Toyota | TPS, JIT, Kaizen, Jidoka |
  74. | [anthropic-researcher](./skills/persona/enterprise/anthropic/anthropic-researcher/SKILL.md) | Anthropic | Constitutional AI, interpretability |
  75. Full list in [CATALOG.md](./CATALOG.md).
  76. ### External ecosystem hub
  77. `external/` is a **registry** of top third-party skill / subagent / plugin repositories —
  78. not a mirror. Pull any subset on demand:
  79. ```bash
  80. python3 scripts/sync_external.py --list # see what's registered
  81. python3 scripts/sync_external.py --all # shallow-clone everything
  82. python3 scripts/sync_external.py --slug anthropics-skills
  83. ```
  84. Registered sources (see [`external/README.md`](./external/README.md) for the full table):
  85. | Category | Repos |
  86. |----------|-------|
  87. | Official (Anthropic) | [`anthropics/skills`](https://github.com/anthropics/skills), [`claude-plugins-official`](https://github.com/anthropics/claude-plugins-official), [`knowledge-work-plugins`](https://github.com/anthropics/knowledge-work-plugins) |
  88. | Curated lists | [`VoltAgent/awesome-agent-skills`](https://github.com/VoltAgent/awesome-agent-skills), [`hesreallyhim/awesome-claude-code`](https://github.com/hesreallyhim/awesome-claude-code), [`travisvn/awesome-claude-skills`](https://github.com/travisvn/awesome-claude-skills), [`ComposioHQ/awesome-claude-skills`](https://github.com/ComposioHQ/awesome-claude-skills) |
  89. | Subagents & orchestration | [`VoltAgent/awesome-claude-code-subagents`](https://github.com/VoltAgent/awesome-claude-code-subagents), [`wshobson/agents`](https://github.com/wshobson/agents), [`0xfurai/claude-code-subagents`](https://github.com/0xfurai/claude-code-subagents) |
  90. To propose a new source, edit [`external/sources.yml`](./external/sources.yml) and open a PR.
  91. External clones are gitignored and never linted by our CI — we are a pointer, not an enforcer.
  92. ### Tooling
  93. The repo ships a Python package for static analysis of skills:
  94. ```bash
  95. pip install -e ./tools/
  96. python -m tools.skill_analyzer.cli score # 8-dimension rubric scores
  97. python -m tools.skill_analyzer.cli tokenizer # Token budget + API cost
  98. python -m tools.skill_analyzer.cli antipattern # Common mistakes scanner
  99. ```
  100. CI (`.github/workflows/quality.yml`) runs these on any PR that touches `skills/`, `tools/`, or `.github/scripts/`, and **blocks merges** whose changed `SKILL.md` files fall below score/token/description thresholds.
  101. ### Known limitations
  102. - **Text-only skills.** Virtually no skill ships executable `scripts/` or `assets/` — they are persona prompts.
  103. - **Description overlap.** The role taxonomy spans 60+ domains and 956 skills and many descriptions overlap, which hurts automatic skill-discovery in agent runtimes. A description-similarity linter runs in CI (informational).
  104. - **Self-scored quality.** `EVALUATION_REPORT.md` files reflect a self-graded rubric, not external review.
  105. - **Three category systems.** `packages/` (14), `roadmap/` (22), and `skills/` (60) use different taxonomies — being consolidated.
  106. ### Documentation
  107. | Doc | Purpose |
  108. |-----|---------|
  109. | [CATALOG.md](./CATALOG.md) | Full catalog of all skills |
  110. | [INSTALL-GUIDE.md](./INSTALL-GUIDE.md) | Platform-specific install instructions |
  111. | [CONTRIBUTING.md](./CONTRIBUTING.md) | How to add or improve skills |
  112. | [.github/CI.md](./.github/CI.md) | CI/CD pipeline notes |
  113. ### License
  114. MIT — see [LICENSE](./LICENSE).
  115. ---
  116. <a name="中文"></a>
  117. ## 中文
  118. ### 项目定位
  119. 按专业领域组织的 **AI Skill 库**,共 **956 个技能**,按 3 种类型(persona / tool / workflow)覆盖 60+ 领域。每个技能是一份 `SKILL.md`(+ 可选 `references/`),本质是**角色化的提示词包**——一套系统指令加参考资料,用于在 Claude Code / OpenCode / Cursor 等 agent 运行时里给 LLM 加载稳定的专业身份、工具专长或工程工作流方法论。
  120. ### 不是什么
  121. - **不是可执行 Skill 套件**。Anthropic/OpenAI 官方 `SKILL.md` 可以随包携带 `scripts/` 和 `assets/` 交给 agent 调用;本仓库绝大多数 skill 仅提供**角色 + 方法论**的文本,应视作有观点的系统提示词,而非可运行插件。
  122. - **不是第三方认证**。仓库内历史质量分由本仓库自己编写的脚本按自家标准打出,对内有参考价值,但不是独立基准。
  123. - **不是每个领域都覆盖完整**。分类很广但深浅不一:有些领域有完整角色树,有些只有一个占位文件。
  124. ### 目录结构
  125. ```
  126. skills/ 956 个技能文件,按 kind 分层组织
  127. persona/ 827 个角色 persona 技能
  128. <domain>/<role>/
  129. SKILL.md Frontmatter + 系统提示词(目标 ≤ 300 行)
  130. references/ 按需加载的深度内容(workflow / scenarios / …)
  131. EVALUATION_REPORT.md 可选:自评质量报告
  132. tool/ 116 个技术工具专家技能(kind: tool)
  133. <technology>/
  134. <skill-name>/SKILL.md
  135. workflow/ 8 个流程驱动工作流技能(kind: workflow)
  136. engineering/ TDD、debug-diagnose、zoom-out、architecture-review、
  137. issue-triage、to-prd
  138. meta/ write-skill、caveman
  139. external/ 精选第三方 skill 仓库的注册表
  140. (anthropics/skills、mattpocock/skills、VoltAgent 等)
  141. 通过 scripts/sync_external.py 按需拉取,不随仓库 vendored
  142. benchmarks/ 评估数据集 + 评分脚本
  143. packages/ 按领域打包的 skill 合集
  144. roadmap/ 职业路径文档(独立于 skills/)
  145. taxonomy.yml 21 个分类、80 个别名的唯一真源
  146. tools/ Skill 分析工具 Python 包(评分、token、反模式)
  147. scripts/ 维护脚本(catalog 重生成、external 同步、taxonomy 检查等)
  148. .github/ CI 工作流与脚本、CI/CD 文档
  149. ```
  150. ### 快速开始
  151. 最通用的安装方式是让 agent 读 SKILL.md 的 URL:
  152. ```
  153. Read https://github.com/theneoai/awesome-skills/blob/main/skills/persona/executive/ceo/SKILL.md 并安装为 skill
  154. ```
  155. 各平台(OpenCode 原生命令、Claude Code、Cursor、Cline、Codex、Kimi)详细步骤见 **[INSTALL-GUIDE.md](./INSTALL-GUIDE.md)**。
  156. ### Skill 合集
  157. 按领域打包,见 `packages/`:
  158. | 合集 | 覆盖 |
  159. |------|------|
  160. | [tech](./packages/tech.md) | 软件、AI/ML、数据 |
  161. | [executive](./packages/executive.md) | CEO、CTO、CFO、COO、CMO |
  162. | [finance](./packages/finance.md) | 银行、咨询、投资 |
  163. | [healthcare](./packages/healthcare.md) | 临床与医疗管理 |
  164. | [software](./packages/software.md) | 后端、前端、devops、QA |
  165. ### 精选企业技能
  166. 仿照特定公司方法论建模的角色(从 `skills/enterprise/` 100+ 条中选 7 条):
  167. | Skill | 公司 | 方法论 |
  168. |-------|------|--------|
  169. | [amazon-engineer](./skills/persona/enterprise/amazon/amazon-engineer/SKILL.md) | Amazon | 14 条领导力准则、Working Backwards、6 页备忘录 |
  170. | [tesla-engineer](./skills/persona/enterprise/tesla/tesla-engineer/SKILL.md) | Tesla | 第一性原理、五步算法 |
  171. | [spacex-engineer](./skills/persona/enterprise/spacex/spacex-engineer/SKILL.md) | SpaceX | 快速迭代、成本创新 |
  172. | [nvidia-ml-engineer](./skills/persona/enterprise/nvidia/nvidia-ml-engineer/SKILL.md) | NVIDIA | CUDA 优化、GPU 平台 |
  173. | [mckinsey-consultant](./skills/persona/enterprise/mckinsey/mckinsey-consultant/SKILL.md) | McKinsey | MECE、Issue Tree、金字塔原理 |
  174. | [toyota-engineer](./skills/persona/enterprise/toyota/toyota-engineer/SKILL.md) | Toyota | TPS、JIT、改善、自働化 |
  175. | [anthropic-researcher](./skills/persona/enterprise/anthropic/anthropic-researcher/SKILL.md) | Anthropic | Constitutional AI、可解释性 |
  176. 完整列表见 [CATALOG.md](./CATALOG.md)。
  177. ### 外部生态集散地
  178. `external/` 是业界优秀 skill / subagent / plugin 仓库的**注册表**(非镜像),按需拉取任意子集:
  179. ```bash
  180. python3 scripts/sync_external.py --list # 查看注册的仓库
  181. python3 scripts/sync_external.py --all # 全量浅拉取
  182. python3 scripts/sync_external.py --slug anthropics-skills
  183. ```
  184. 已注册仓库(完整表格见 [`external/README.md`](./external/README.md)):
  185. | 类别 | 仓库 |
  186. |------|------|
  187. | 官方(Anthropic) | [`anthropics/skills`](https://github.com/anthropics/skills)、[`claude-plugins-official`](https://github.com/anthropics/claude-plugins-official)、[`knowledge-work-plugins`](https://github.com/anthropics/knowledge-work-plugins) |
  188. | 社区精选列表 | [`VoltAgent/awesome-agent-skills`](https://github.com/VoltAgent/awesome-agent-skills)、[`hesreallyhim/awesome-claude-code`](https://github.com/hesreallyhim/awesome-claude-code)、[`travisvn/awesome-claude-skills`](https://github.com/travisvn/awesome-claude-skills)、[`ComposioHQ/awesome-claude-skills`](https://github.com/ComposioHQ/awesome-claude-skills) |
  189. | Subagent 与编排 | [`VoltAgent/awesome-claude-code-subagents`](https://github.com/VoltAgent/awesome-claude-code-subagents)、[`wshobson/agents`](https://github.com/wshobson/agents)、[`0xfurai/claude-code-subagents`](https://github.com/0xfurai/claude-code-subagents) |
  190. 提议新增来源:编辑 [`external/sources.yml`](./external/sources.yml) 并提交 PR。
  191. `external/` 下的本地克隆已加入 `.gitignore`,也不被 CI 扫描——我们是**指针**,不是上游的审阅者。
  192. ### 工具
  193. 仓库附带 Python 分析包:
  194. ```bash
  195. pip install -e ./tools/
  196. python -m tools.skill_analyzer.cli score # 8 维度评分
  197. python -m tools.skill_analyzer.cli tokenizer # Token 预算与 API 成本
  198. python -m tools.skill_analyzer.cli antipattern # 反模式扫描
  199. ```
  200. CI(`.github/workflows/quality.yml`)会在修改了 `skills/`、`tools/` 或 `.github/scripts/` 的 PR 上运行这些工具,若新改动的 `SKILL.md` 达不到阈值则**阻塞合并**。
  201. ### 已知局限
  202. - **绝大多数 skill 是纯文本 persona**,并不随包携带可执行 `scripts/` 或 `assets/`。
  203. - **描述重叠严重**:60+ 分类 × 956 个技能,许多 description 字段相似度高,会降低 agent 运行时的 skill 自动发现准确度。描述相似度检查已在 CI 中运行(仅提示,不阻塞)。
  204. - **质量分是自评**:`EVALUATION_REPORT.md` 反映仓库自家规则的评分,不是独立评审。
  205. - **三套分类互不一致**:`packages/`(14 类)、`roadmap/`(22 类)、`skills/`(60 类)仍在统一中。
  206. ### 文档
  207. | 文档 | 用途 |
  208. |------|------|
  209. | [CATALOG.md](./CATALOG.md) | 完整技能目录 |
  210. | [INSTALL-GUIDE.md](./INSTALL-GUIDE.md) | 各平台安装步骤 |
  211. | [CONTRIBUTING.md](./CONTRIBUTING.md) | 如何贡献或改进技能 |
  212. | [.github/CI.md](./.github/CI.md) | CI/CD 说明 |
  213. ### 相关项目
  214. - [Awesome MCPs](https://theneoai.github.io/awesome-mcps/) - 115+ MCP 服务器,一键安装
  215. - [GitHub](https://github.com/theneoai/awesome-mcps) - Awesome MCPs 项目源码
  216. ### 许可证
  217. MIT,见 [LICENSE](./LICENSE)。
  218. ---
  219. <div align="center">
  220. **[↑ Back to top](#awesome-skills)**
  221. </div>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.