# Anthropic Cybersecurity Skills — 754 Skills Library

> Anthropic Cybersecurity Skills is a 754-skill library for AI agents, mapped to 5 frameworks (ATT&CK, NIST CSF 2.0, ATLAS, D3FEND, AI RMF).

## Install

Save the content below to `.claude/skills/` or append to your `CLAUDE.md`:

## Quick Use

```bash
npx skills add mukul975/Anthropic-Cybersecurity-Skills
# Or clone:
git clone https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git
```

## Intro

Anthropic Cybersecurity Skills is a 754-skill library for AI agents, mapped to 5 frameworks (ATT&CK, NIST CSF 2.0, ATLAS, D3FEND, AI RMF).

**Best for:** DevSecOps teams who want structured incident response and hunting playbooks

**Works with:** Claude Code, Codex CLI, Cursor, Gemini CLI, agentskills.io-style skill runners

**Setup time:** 3-8 minutes

### Key facts (verified)

- GitHub: 6258 stars · 866 forks · pushed 2026-05-13.
- License: Apache-2.0 · owner avatar + repo URL verified via GitHub API.
- README-verified entrypoint: `npx skills add mukul975/Anthropic-Cybersecurity-Skills`.

## Main

- Use it as a skills corpus: point your agent to the repo’s skill directory, then search by domain (e.g., DFIR, cloud, web) instead of free-text prompting.

- Leverage the five-framework mapping when you need to explain “why this step exists” (ATT&CK/CSF/ATLAS/D3FEND/AI RMF) to auditors or security leads.

- Start with a narrow workflow (one incident type or one cloud provider) and only then broaden to the full 26-domain library to keep agent planning focused.

### Source-backed notes

- README states the repo contains 754 structured cybersecurity skills across 26 domains and follows the agentskills.io standard.
- README includes a “Quick start” with `npx skills add ...` and a git-clone alternative.
- README notes it is a community project and explicitly not affiliated with Anthropic PBC.

### FAQ

- **Is it officially from Anthropic?**: No — README labels it a community project and not affiliated with Anthropic PBC.
- **How do I install it fastest?**: Use the README quick start: `npx skills add mukul975/Anthropic-Cybersecurity-Skills`.
- **Do I need to run any binaries?**: Not necessarily; it’s primarily a skills library you attach to your agent and navigate via search.

## Source & Thanks

> Source: https://github.com/mukul975/Anthropic-Cybersecurity-Skills
> License: Apache-2.0
> GitHub stars: 6258 · forks: 866

---

<!-- ZH -->

## Quick Use

```bash
npx skills add mukul975/Anthropic-Cybersecurity-Skills
# Or clone:
git clone https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git
```

## Intro

Anthropic Cybersecurity Skills 是面向 AI agent 的 754 条网络安全技能库，并映射到 5 个框架（MITRE ATT&CK、NIST CSF 2.0、ATLAS、D3FEND、AI RMF）。用 npx 一键装到 skills 后即可检索与执行。

**Best for:** 想把 IR/狩猎流程变成可检索技能库的 DevSecOps 团队

**Works with:** Claude Code、Codex CLI、Cursor、Gemini CLI，以及 agentskills.io 风格 skill runner

**Setup time:** 3-8 minutes

### Key facts (verified)

- GitHub：6258 stars · 866 forks；最近更新 2026-05-13。
- 许可证：Apache-2.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中核对过的入口命令：`npx skills add mukul975/Anthropic-Cybersecurity-Skills`。

## Main

- 把它当作“技能语料库”：让 agent 面向技能目录检索（DFIR/云/应用安全等），而不是靠自由文本乱猜下一步。

- 需要对齐合规/审计时，用 5 框架映射解释“为什么做这一步”（ATT&CK/CSF/ATLAS/D3FEND/AI RMF）。

- 先从单一事件类型或单一云厂商的流程开始，验证可用后再扩到 26 个域，避免 agent 规划发散。

### Source-backed notes

- README 写明：该库包含 754 条结构化安全技能、覆盖 26 个安全域，并遵循 agentskills.io 标准。
- README 的 Quick start 提供 `npx skills add ...`，也给出 git clone 方案。
- README 明确声明：社区项目，与 Anthropic PBC 无官方关联。

### FAQ

- **这是 Anthropic 官方出品吗？**：不是。README 标注为社区项目，并声明与 Anthropic PBC 无关联。
- **最快怎么安装？**：按 README Quick start：`npx skills add mukul975/Anthropic-Cybersecurity-Skills`。
- **需要跑什么可执行程序吗？**：通常不需要，它主要是技能库；把 skills 挂到 agent 后按主题检索/执行即可。

## Source & Thanks

> Source: https://github.com/mukul975/Anthropic-Cybersecurity-Skills
> License: Apache-2.0
> GitHub stars: 6258 · forks: 866


---
Source: https://tokrepo.com/en/workflows/anthropic-cybersecurity-skills-754-skills-library
Author: Skill Factory