SunnyKumar-Jonwal/Anthropic-Cybersecurity-Skills ? reverse-engineered prompt

Build me an open source cybersecurity skills library for AI agents that I can drop into tools like Claude Code, Cursor, Copilot, and similar apps.

I want a big collection of structured security skills, each one written in a consistent format and mapped to the right industry frameworks like MITRE ATT&CK, NIST CSF, D3FEND, ATLAS, NIST AI RMF, and MITRE Fight Fraud. It should cover a lot of common security work like detection, incident response, forensics, cloud security, threat hunting, and fraud related investigations, with clear metadata so an agent can pick the right skill fast.

Please include the docs, the skill files, the mapping files, and a simple way for an AI agent to load and use them. Keep it practical and production ready, with good README docs and examples. If you need to check current framework details or format expectations, look them up online first.