Anthropic-Cybersecurity-Skills - 817 structured cybersecurity skills for AI agents

项目说明

<p align="center">
<img src="assets/banner.png" alt="Anthropic Cybersecurity Skills" width="100%">
</p>

<di="center">

Anthropic Cybersecurity Skills

The largest open-source cybersecurity skills library for AI agents

![GARS-2026 Survey]()
![MITRE F3]()
![GitHub stars]()
![GitHub forks]()
![Last Commit]()
![agentskills.io]()
![Playground]()
![Hermes Agent]()

818 production-grade cybersecurity skills · 34 security domains · 6 framework mappings · 26+ AI platforms

Get Started · What's Inside · Frameworks · Platforms · Contributing

</div>

---

> ⚠️ Community Project — This is an independent, community-created project. Not affiliated with Anthropic PBC.
>
> 🔐 Authorized & lawful use only. This library includes offensive and dual-use techniques (e.g. red-team C2, phishing simulation, exploitation) intended for authorized penetration testing, security research, defense, and education. Only use them against systems you own or have explicit written permission to test, and comply with all applicable laws and rules of engagement. You are solely responsible for how you use these skills. See SECURITY.md and CODE_OF_CONDUCT.md.

Give any AI agent the security skills of a senior analyst

A junior analyst knows which plugin to run on a suspicious memory dump, which Sigma rules catch Kerberoasting, and how to scope a cloud breach across three pro. Your AI agent doesn't — unless you give it these skills.

This repo contains 818 structured cybersecurity skills spanning 34 security domains, each following the agentskills.io open standard. The library maps across six industry frameworks — MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, MITRE D3FEND, NIST AI RMF, and the MITRE Fight Fraud Framework (F3) — with each skill mapped to the frameworks relevant to its type (a forensics skill carries ATT&CK + CSF; an AI-security skill adds ATLAS and AI RMF). Clone it, point your agent at it, and your next security in gets expert-le in seconds.

Six frameworks, one skill library

Each skill maps to the frameworks that fit its subject — ATT&CK and NIST CSF are near-uni, while ATLAS, AI RMF, D3FEND, and F3 apply where they're relevant. Framework co across the 817 skills: MITRE ATT&CK 805 · NIST CSF 2.0 804 · MITRE D3FEND 139 · NIST AI RMF 97 · MITRE F3 94 · MITRE ATLAS 93.

| Framework | | Framework scope | What it maps |
|---|---|---|---|
| MITRE ATT&CK | v19.1 | 15 tactics · Enterprise/Mobile/ICS | Ad behaviors and TTPs |
| NIST CSF 2.0 | 2.0 | 6 functions · 22 categories · 106 subcategories | Organizational security posture |
| MITRE ATLAS | 2026.07 | 101 techniques · 77 sub-techniques | AI/ML ad threats |
| MITRE D3FEND | v1.4.0 | 270 techniques | Defensi |
| NIST AI RMF | 1.0 | 4 functions (Govern/Map/Measure/Manage) | AI risk management |
| MITRE F3 (Fight Fraud Framework) | v1.1 (2026-04-09) | 8 tactics · 123 techniques · 94 fraud-relevant skills | Cyber-enabled financial fraud TTPs |

Example — each skill maps only to the frameworks relevant to it (one may hit all six, another just a couple):

| Skill | ATT&CK | NIST CSF | ATLAS | D3FEND | AI RMF | F3 |
|---|---|---|---|---|---|---|
| analyzing-network-traffic-of-malware | T1071 | DE.CM | AML.T0047 | D3-NTA | MEASURE-2.6 | — |
| detecting-business-email-compromise | T1566 | DE.AE | — | — | — | F1005.006 · monetization |

🆕 MITRE Fight Fraud Framework (F3) — 94 fraud-relevant skills

![MITRE F3]()

The MITRE Fight Fraud Framework (F3) was released April 9, 2026 by MITRE's Center for Threat-Informed Defense (CTID), co-de with JPMorganChase, Citigroup, Lloyds Banking Group, Standard Chartered, CrowdStrike, Business, FS-ISAC, and others. It is an ATT&CK-compatible TTP catalog for cyber-enabled financial fraud — filling the gap ATT&CK lea initial compromise.

F3 v1.1 adds two fraud-specific tactics that ATT&CK does not enumerate:

Positioning (FA0001) — actions taken after access to collect/manipulate data and prepare the fraud (synthetic-identity seeding, account warming, beneficiary setup, SIM-swap pre-positioning, banking-session hijack).

Monetization (FA0002) — con stolen assets into usable funds (money-mule layering, APP fraud, crypto off-ramping, card cash-out, refund/chargeback abuse).

Fraud-specific techniques use F1XXX IDs (e.g. F1005.003 Add Beneficiary, F1025.003 Wire Transfer, F1007 Ad); reused ATT&CK techniques keep their T1XXX IDs. Mappings live in each skill's mitre_f3: frontmatter block — all 123 F3 v1.1 technique IDs were against the upstream STIX bundle. See docs/mitre-f3-mapping.md for the schema.

MITRE ATT&CK v19.1 — 805/817 skills mapped

Every skill carries a mitre_attack frontmatter list against MITRE ATT&CK v19.1 (the latest release) using the official mitreattack-python library — 290 distinct techniques and sub-techniques (146 base + 144 sub) across Enterprise, ICS, and Mobile. Zero revoked or deprecated IDs. v19.1's restructured Defense E (now split into Stealth and Defense Impairment) is reflected below.

| Tactic | ID | Skills |
|--------|----|--------|
| Reconnaissance | TA0043 | 103 |
| Resource De | TA0042 | 22 |
| Initial Access | TA0001 | 467 |
| Execution | TA0002 | 350 |
| Persistence | TA0003 | 444 |
| Pri Escalation | TA0004 | 464 |
| Stealth | TA0005 | 442 |
| Defense Impairment | TA0112 | 92 |
| Credential Access | TA0006 | 202 |
| Discovery | TA0007 | 237 |
| Lateral Mo | TA0008 | 68 |
| Collection | TA0009 | 172 |
| Command and Control | TA0011 | 123 |
| Exfiltration | TA0010 | 82 |
| Impact | TA0040 | 50 |

Quick start

Works immediately with Claude Code, GitHub Copilot, OpenAI Codex CLI, Cursor, Gemini CLI, and any agentskills.io-compatible platform.

🌍 GARS-2026 — Global Agentic AI Readiness Survey

I'm running a global academic study measuring how ready security professionals,
de, and enterprise teams actually are for agentic AI — MCP servers,
tool calling, go, and human-in-the-loop workflows.

If you use this repo, your response would be a genuinely data point.

📋 Take the survey (10 min):
Survey Link

60 questions · Anonymous · Supervised by SRH Berlin

You get 50 Casky Tokens for early access to casky.ai

Results published open access under CC-BY 4.0

🚀 Try it on the Playground

Experience Casky.ai hands-on — no setup required.

→ Launch Playground on Casky.ai

The playground lets you:

Run li skill exercises against real targets

See AI agents execute structured skills in real time

Explore MITRE ATT&CK mapped workflows interactively

Test threat hunting, DFIR, and penetration testing scenarios

No installation. No configuration. Just open and start.

Why this exists

The cybersecurity workforce gap hit 4.8 million unfilled roles globally in 2024 (ISC2). AI agents can help close that gap — but only if they ha domain knowledge to work from. Today's agents can write code and search the web, but they lack the practitioner playbooks that turn a generic LLM into a capable security analyst.

Existing security tool repos give you wordlists, payloads, or exploit code. None of them give an AI agent the structured decision-making workflow a senior analyst follows: when to use each technique, what prerequisites to check, how to execute step-by-step, and how to results. That is the gap this project fills.

Anthropic Cybersecurity Skills is not a collection of scripts or checklists. It is an AI-nati base built from the ground up for the agentskills.io standard — YAML frontmatter for sub-second discovery, structured Markdown for step-by-step execution, and reference files for deep technical context. Every skill encodes real practitioner workflows, not generated summaries.

What's inside — 34 security domains

| Domain | Skills | Key capabilities |
|---|---|---|
| Cloud Security | 66 | AWS, Azure, GCP hardening · CSPM · cloud attack emulation · cloud forensics |
| SOC Operations | 63 | Playbooks · escalation workflows · Graph-log detection · tabletop exercises |
| Threat Hunting | 58 | Hypothesis-dri · LOTL detection · E · fleet hunting |
| Threat Intelligence | 52 | STIX/TAXII · MISP · OpenCTI · feed integration · actor profiling |
| Web Application Security | 46 | OWASP Top 10 · SQLi · XSS · SSRF · deserialization |
| Network Security | 43 | IDS/IPS · firewall rules · VLAN segmentation · traffic analysis |
| Digital Forensics | 41 | Disk imaging · memory forensics · Hayabusa/KAPE/Plaso timelines |
| Identity & Access Management | 40 | Entra ID/ROADtools · de phishing · PAM · zero trust identity |
| Malware Analysis | 39 | Static/dynamic analysis · reverse engineering · sandboxing |
| Red Teaming | 35 | ADCS/Certipy · BloodHound CE · Sli C2 · NTLM relay |
| Container Security | 33 | K8s RBAC · image scanning · Falco · container escape |
| OT/ICS Security | 29 | Modbus · DNP3 · IEC 62443 · historian defense · SCADA |
| API Security | 28 | GraphQL · REST · OWASP API Top 10 · WAF bypass |
| Incident Response | 26 | Breach containment · ransomware response · IR playbooks |
| Management | 25 | Nessus · scanning workflows · patch prioritization · CVSS |
| Penetration Testing | 23 | Network · web · cloud · mobile · NetExec lateral mo |
| De | 18 | CI/CD security · Trivy IaC/image scanning · code signing |
| Zero Trust Architecture | 18 | BeyondCorp · CISA maturity model · microsegmentation |
| Endpoint Security | 17 | EDR · LOTL detection · fileless malware · persistence hunting |
| Phishing Defense | 16 | Email authentication · BEC detection · phishing IR |
| Cryptography | 16 | TLS · Ed25519 · post-quantum migration · key management |
| AI Security | 14 | LLM red-teaming (garak/PyRIT) · prompt injection · MCP/agentic security · guardrails |
| Mobile Security | 13 | Android/iOS analysis · mobile pentesting · MDM forensics |
| Ransomware Defense | 13 | Precursor detection · response · recovery · encryption analysis |
| Compliance & Go | 10 | NIST 800-30/RMF · CMMC · HIPAA · TPRM · CIS benchmarks |
| Supply Chain Security | 8 | SBOMs · dependency confusion · malicious-package triage · SLSA/Sigstore |
| Threat Detection | 7 | Credential dumping · golden-ticket forgery · pass-the-ticket · LOLBAS · UEBA insider signals |
| Hardware & Firmware Security | 6 | CHIPSEC/UEFI audit · Secure Boot bypass · TPM attestation · bootkit hunting |
| Deception Technology | 6 | Honeytokens · canarytokens · breach detection |
| Blockchain Security | 2 | Ethereum smart-contract · Foundry audit workflows |
| Wireless Security | 2 | Bluetooth Low Energy attack detection · BLE security assessment |
| Privacy Compliance | 2 | GDPR data-subject access requests · privacy impact assessments |
| Data Protection | 1 | Data loss pre with Microsoft Purview |
| Purple Team | 1 | Atomic Red Team purple-team testing |

*817 skills across 34 domains. Counts come from the subdomain field in each skill's frontmatter.*

How AI agents use these skills

Each skill costs ~30 tokens to scan (frontmatter only) and 500–2,000 tokens to fully load (complete workflow). This progressi architecture lets agents search all 818 skills in a single pass without blowing context windows.

Without these skills, the agent guesses at tool commands and misses critical steps. With them, it follows the same playbook a senior DFIR analyst would use.

Skill anatomy

Every skill follows a consistent directory structure:

YAML frontmatter (real example)

Markdown body sections

Frontmatter fields: name (kebab-case, 1–64 chars), description (keyword-rich for agent discovery), domain, subdomain, tags, atlas_techniques (MITRE ATLAS IDs), d3fend_techniques (MITRE D3FEND IDs), nist_ai_rmf (NIST AI RMF references), nist_csf (NIST CSF 2.0 categories). MITRE ATT&CK technique mappings are documented in each skill's references/standards.md file and in the ATT&CK Na layer included with releases.

<details>
<summary><strong>📊 MITRE ATT&CK Enterprise co — all 15 tactics</strong></summary>

&nbsp;

| Tactic | ID | Co | Key skills |
|---|---|---|---|
| Reconnaissance | TA0043 | Strong | OSINT, subdomain enumeration, DNS recon |
| Resource De | TA0042 | Moderate | Phishing infrastructure, C2 setup detection |
| Initial Access | TA0001 | Strong | Phishing simulation, exploit detection, forced browsing |
| Execution | TA0002 | Strong | PowerShell analysis, fileless malware, script block logging |
| Persistence | TA0003 | Strong | Scheduled tasks, registry, service accounts, LOTL |
| Pri Escalation | TA0004 | Strong | Kerberoasting, AD attacks, cloud pri escalation |
| Stealth | TA0005 | Strong | Obfuscation, rootkit analysis, e detection |
| Defense Impairment | TA0112 | Moderate | Impair Defenses (T1562), log/indicator removal, EDR tampering |
| Credential Access | TA0006 | Strong | Mimikatz detection, pass-the-hash, credential dumping |
| Discovery | TA0007 | Moderate | BloodHound, AD enumeration, network scanning |
| Lateral Mo | TA0008 | Strong | SMB exploits, lateral mo detection with Splunk |
| Collection | TA0009 | Moderate | Email forensics, data staging detection |
| Command and Control | TA0011 | Strong | C2 beaconing, DNS tunneling, Cobalt Strike analysis |
| Exfiltration | TA0010 | Strong | DNS exfiltration, DLP controls, data loss detection |
| Impact | TA0040 | Strong | Ransomware defense, encryption analysis, recovery |

An ATT&CK Na layer file is included in the v1.0.0 release assets for co mapping.

> Note: ATT&CK v19 lands April 28, 2026 — splitting Defense E (TA0005) into two new tactics: *Stealth* and *Impair Defenses*. Skill mappings will be updated in a forthcoming release.

</details>

<details>
<summary><strong>📊 NIST CSF 2.0 alignment — all 6 functions</strong></summary>

&nbsp;

| Function | Skills | Examples |
|---|---|---|
| Govern (GV) | 30+ | Risk strategy, policy frameworks, roles & responsibilities |
| Identify (ID) | 120+ | Asset discovery, threat landscape assessment, risk analysis |
| Protect (PR) | 150+ | IAM hardening, WAF rules, zero trust, encryption |
| Detect (DE) | 200+ | Threat hunting, SIEM correlation, anomaly detection |
| Respond (RS) | 160+ | Incident response, forensics, breach containment |
| Recover (RC) | 40+ | Ransomware recovery, BCP, disaster recovery |

NIST CSF 2.0 (February 2024) added the Govern function and expanded scope from critical infrastructure to all organizations. Skill mappings align to all 22 categories and reference 106 subcategories.

</details>

<details>
<summary><strong>📊 Framework deep dive — ATLAS, D3FEND, AI RMF</strong></summary>

&nbsp;

MITRE ATLAS 2026.07 — AI/ML ad threats

ATLAS maps ad tactics, techniques, and case studies specific to AI and machine learning systems. Release 2026.07 covers 101 techniques and 77 sub-techniques including agentic AI attack added in late 2025: AI agent context poisoning, tool in abuse, MCP ser, and malicious agent deployment. Skills mapped to ATLAS help agents identify and defend against threats to ML pipelines, model weights, inference APIs, and autonomous workflows.

MITRE D3FEND v1.4.0 — Defensi

D3FEND is an NSA-funded knowledge graph of 270 defensi organized across 7 tactical categories: Model, Harden, Detect, Isolate, Decei, and Restore. Built on OWL 2 ontology, it uses a shared Digital Artifact layer to bidirectionally map defensi to ATT&CK offensi. Skills tagged with D3FEND identifiers let agents recommend specific countermeasures for detected threats.

NIST AI RMF 1.0 + GenAI Profile (AI 600-1)

The AI Risk Management Framework defines 4 core functions — Govern, Map, Measure, Manage — with 72 subcategories for trustworthy AI de. The GenAI Profile (AI 600-1, July 2024) adds 12 risk categories specific to generative AI, from confabulation and data privacy to prompt injection and supply chain risks. Colorado's AI Act (effecti 2026) provides a legal safe harbor for organizations complying with NIST AI RMF, making these mappings directly relevant to regulatory compliance.

</details>

Compatible platforms

AI code assistants
Claude Code (Anthropic) · GitHub Copilot (Microsoft) · Cursor · Windsurf · Cline · Aider · Continue · Roo Code · Amazon Q De · Tabnine · Sourcegraph Cody · JetBrains AI

CLI agents
OpenAI Codex CLI · Gemini CLI (Google)

Autonomous agents
Devin · Replit Agent · SWE-agent · OpenHands

Agent frameworks & SDKs
LangChain · CrewAI · AutoGen · Semantic Kernel · Haystack · AI SDK · Any MCP-compatible agent

All platforms that support the agentskills.io standard can load these skills with zero configuration.

What people are saying

> *"A database of real, organized security skills that any AI agent can plug into and use. Not tutorials. Not blog posts."*
> — Hasan Toor (), AI/tech creator

> *"This is not a random collection of security scripts. It's a structured operational knowledge base designed for AI-dri workflows."*
> — fazal-sec, Medium

Featured in

| Where | Type | Link |
|---|---|---|
| awesome-agent-skills | Awesome List (1,000+ skills index) | /awesome-agent-skills |
| awesome-ai-security | Awesome List (AI security tools) | ottosulin/awesome-ai-security |
| awesome-codex-cli | Awesome List (Codex CLI resources) | RoggeOhta/awesome-codex-cli |
| SkillsLLM | Skills directory & marketplace | skillsllm.com/skill/anthropic-cybersecurity-skills |
| Openflows | Signal analysis & tracking | openflows.org |
| Ne skills_feed | Automated skills index | Ne/skills_feed |

Star history

<a href="
<picture>
<source media="(prefers-color-scheme: dark)" srcset=" />
<source media="(prefers-color-scheme: light)" srcset=" />
<img alt="Star History Chart" src=" width="100%" />
</picture>
</a>

Releases

| | Date | Highlights |
|---|---|---|
| v1.0.0 | March 11, 2026 | 734 skills · 26 domains · MITRE ATT&CK + NIST CSF 2.0 mapping · ATT&CK Na layer |

Skills ha to grow on main since v1.0.0 — the library now contains 818 skills with 6-framework mapping (MITRE ATLAS, D3FEND, NIST AI RMF, and the MITRE Fight Fraud Framework added post-release). Check Releases for the latest tagged .

Contributing

This project grows through community contributions. Here is how to get in:

Add a new skill — Read SCOPE.md first, then the template in CONTRIBUTING.md. The thinnest domains are the ones most worth adding to: Data Protection and Purple Team have one skill each, and Blockchain Security, Wireless Security and Privacy Compliance have two. Submit one skill per PR, with the title Add skill: your-skill-name.

Impro skills — Add framework mappings, fix workflows, update tool references, or contribute scripts and templates. Deepening a skill that already exists is worth as much to this library as adding a new one.

Report issues — Found an inaccurate procedure or broken script? Open an issue.

Every PR is re for technical accuracy and agentskills.io standard compliance. I do that review myself, and the queue is currently longer than I would like — some pull requests have been open for months. I am working through them, and small, focused PRs mo. Check good first issues for a starting point.

This project follows the Contributor Co. By participating, you agree to uphold this code.

🙏 Thanks to our contributors

This library is built by the community. Thank you to e who has contributed:

<!-- contributors:start -->
<p align="center">
<a href=" title="mukul975 — maintainer"><img src=" width="72" height="72" alt=""></a>
<a href=" title=" — 18 contributions"><img src=" width="72" height="72" alt="@"></a>
<a href=" title="juliosuas — 13 contributions"><img src=" width="72" height="72" alt=""></a>
<a href=" title="Daytona39264 — 3 contributions"><img src=" width="72" height="72" alt=""></a>
<a href=" title="ke — 2 contributions"><img src=" width="72" height="72" alt="@ke"></a>
<a href=" title="andrewibrah — 2 contributions"><img src=" width="72" height="72" alt=""></a>
<a href=" title="Bortlesboat — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="De — 1 contribution"><img src=" width="72" height="72" alt="@De"></a>
<a href=" title="ioxoi — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="OctoBored — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="shanujans — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="farhan6667 — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="daksh — 1 contribution"><img src=" width="72" height="72" alt=""></a>
<a href=" title="nyxst4ck — 1 contribution"><img src=" width="72" height="72" alt=""></a>
</p>

<p align="center">_{14 contributors, ordered by contribution count · see the full <a href=" graph</a>}</p>
<!-- contributors:end -->

Community

💬 Discussions — Questions, ideas, and roadmap con
🐛 Issues — Bug reports and feature requests
🔒 Security Policy — Responsible disclosure process (48-hour acknowledgment)

Citation

If you use this project in research or publications:

License

This project is licensed under the Apache License 2.0. You are free to use, modify, and distribute these skills in both personal and commercial projects.

---

<di="center">

If this project helps your security work, consider giving it a ⭐

⭐ Star · 🍴 Fork · 💬 Discuss · 📝 Contribute

Community project by . Not affiliated with Anthropic PBC.

</div>

项目信息

开发语言：Python

Star 数：33797

技术要点

该项目属于病毒分析方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕817 structured cybersecurity skills for AI agents · Mapped to 6 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, NIST AI RMF & MITRE F3 (Fight Fraud) · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platforms · 29 security domains · Apache 2.0展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。