Anthropic-Cybersecurity-Skills: 817 Skills: Coding Senior Analyst Judgment into Agents

An open-source knowledge base designed to bridge the judgment gap caused by a shortage of security personnel. It provides structured workflows that agents can execute immediately, rather than simple scripts or checklists. From selecting Volatility3 plugins to determining the scope of cloud breaches, it codifies the decision-making process taken by senior analysts.

It includes 817 skills across 29 security domains, with each skill following the agentskills.io standard. Mapped to six major frameworks including MITRE ATT&CK v19.1, NIST CSF 2.0, and MITRE ATLAS, it allows for instant retrieval of relevant response procedures and verification steps by inputting specific technique IDs. Notably, it also supports the 94 techniques of the MITRE F3 (Financial Fraud Response Framework) released in April 2026.

Unlike existing security tool repositories that only provide word lists or exploit code, this library contains practical playbooks explaining the 'why' and 'how'. Stored in YAML front matter and structured Markdown formats, it enables AI agents to discover and execute required skills within milliseconds. It works immediately with over 26 AI platforms, including Claude Code, GitHub Copilot, and Cursor.

Use is permitted only for authorized penetration testing, security research, defense, and educational purposes. As it includes offensive techniques and dual-use technologies, it must be applied only to systems you own or have explicit written permission to test. Legal compliance and liability rest with the user; relevant policies can be found in SECURITY.md.