Mukul975 releases a comprehensive set of 754 structured cybersecurity skills mapped to major industry frameworks for AI agents.
Anthropic-Cybersecurity-Skills is an open-source repository containing 754 structured, AI-native cybersecurity skills. Following the agentskills.io standard, each skill mimics the decision-making process of a senior analyst by defining prerequisites, step-by-step execution guides, and verification methods in YAML and Markdown format. These skills are mapped to five leading industry frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, and NIST AI RMF. They are fully compatible with over 20 agent platforms and tools, including Claude Code, GitHub Copilot, Codex CLI, Cursor, and Gemini CLI.
Converting passive cybersecurity checklists into active, machine-executable skills is a game-changer that will accelerate the adoption of autonomous AI security analysts.
* **Bridges the Execution Gap:** Shifts the focus from theoretical compliance to automated, tactical execution by giving AI agents step-by-step instructions across 26 security domains.
* **Standardized Portability:** Built on the emerging agentskills.io standard, ensuring that security operations can easily be ported across different AI platforms (e.g., Claude Code, Cursor, Gemini CLI) without vendor lock-in.
* **Robust Alignment:** By mapping to established frameworks like MITRE ATT&CK and NIST CSF 2.0, it allows organizations to directly translate security frameworks into automated agent workflows.
DISCOVERED
3h ago
2026-06-22
PUBLISHED
3h ago
2026-06-22
RELEVANCE