mukul975/Anthropic-Cybersecurity-Skills
An open-source library that gives AI agents 817 ready-made cybersecurity playbooks
This repository packages 817 structured 'skill' documents that AI coding tools and agents can reference when doing cybersecurity work. Each skill spells out when to use which tool and what steps to follow, and is mapped to six industry frameworks including MITRE ATT&CK. It plugs directly into 20+ AI platforms such as Claude Code, GitHub Copilot, and Cursor.
What it does
- 817 skill documents are organized across 29 security domains, covering things like memory forensics, phishing detection, and cloud breach response
- Each skill has a YAML summary block plus a markdown body with 'When to Use, Prerequisites, Workflow, Verification' sections, letting agents first scan lightweight summaries (~30 tokens each) and only fully load (500-2,000 tokens) the ones they need
- Skills are mapped to whichever of six frameworks fit them: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, MITRE D3FEND, NIST AI RMF, and MITRE F3 (fraud framework)
- Installable via npx or git clone, and works out of the box with any agentskills.io-compatible platform, including Claude Code, Copilot, Cursor, and Gemini CLI
- Released under Apache 2.0 as a community project; it explicitly states it is not affiliated with Anthropic
Why it matters
Amid a shortage of security professionals, this gives AI agents a practical, procedural knowledge base to follow real analyst workflows instead of guessing. The repo itself warns that it includes offensive techniques, so it must only be used with proper authorization.
Terms in this repo
- MITRE ATT&CK · an industry-standard knowledge base cataloging real-world attacker tactics and techniques
- NIST CSF · a U.S. NIST framework for assessing an organization's overall security posture
- agentskills.io · an open standard defining how AI agents should read and use skill documents
- YAML frontmatter · a metadata block at the top of a document summarizing tags and description
- D3FEND · a MITRE knowledge base of defensive countermeasures mapped against attack techniques
Repository description (English)
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
Open on GitHubTrending repos
- cathrynlavery/diagram-designA skill that makes AI coding tools draw magazine-quality diagrams instead of generic rounded boxes
- public-apis/public-apisA giant crowd-curated directory of free APIs for developers
- semantica-agi/semanticaAn open-source graph infrastructure that lets AI agents show their work, not just their answers
- cactus-compute/needleA 14MB AI model small enough to run tool-calling on a phone or watch, without internet
- unslothai/unslothA desktop app that lets you run and train AI models on your own computer, no coding required
- macro-inc/macroAn all-in-one workspace where email, chat, docs, tasks, and CRM are cross-linked and share one AI memory
- harry0703/MoneyPrinterTurboAn open-source tool that turns a single topic or keyword into a finished short video, complete with script, footage, subtitles, and music
- basecamp/omarchyA ready-made, opinionated Linux setup built by DHH
Latest from METAL LAB
- Grok leaks conversations, location data via commands hidden in ciphertext
- Cursor lets agents handle long-running tasks with new "/goal" command
- NVIDIA releases Switchyard, an LLM routing proxy for coding agents
- Caveman Cuts Claude Code Token Usage by 33% Using Caveman-Speak
- Apple Keeps 2027 Target for Camera AirPods, Qwen3.8-27B Makes 80 Tool Calls Locally