Библиотека скиллов по кибербезопасности для AI-агентов

general

Anthropic-Cybersecurity-Skills is an open-source library of 817 production-grade cybersecurity skills for Claude Code and other AI agents, spanning 29 security domains including forensics, malware analysis, penetration testing, cloud security, fraud detection, and compliance. Each skill follows the agentskills.io open standard and is mapped to relevant industry frameworks: MITRE ATT&CK v19.1 (805 skills), NIST CSF 2.0 (804), MITRE D3FEND (139), NIST AI RMF (97), MITRE ATLAS (93), and MITRE F3 v1.1 (94 fraud-focused skills). Compatible with 26+ AI platforms, the library covers both defensive and dual-use offensive techniques intended solely for authorized penetration testing, security research, and education. Security engineers and analysts use it to equip AI agents with senior-analyst-level guidance — from Volatility3 memory dump analysis and Sigma rule tuning to multi-cloud breach scoping.

• #abusing-dpapi-for-credential-access
• #analyzing-memory-dumps-with-volatility
• #analyzing-malware-behavior-with-cuckoo-s
• #analyzing-network-traffic-with-wireshark
• #achieving-cmmc-level-2-compliance