NVISOsecurity/cyber-security-llm-agents — reverse-engineered prompt

Build me a Python based cyber security LLM agent toolkit for a lab environment. I want security analysts to be able to configure their LLM API keys in an env file, install requirements, and run named scenarios from the command line, starting with a simple hello agents demo that proves the setup works.

The project should have modular agents and reusable tasks so people can combine them into different workflows for common security work, like analysis, adversary emulation demos, detection engineering experiments, and purple team style automation. Keep it easy to add new scenarios in one central place and run them by name.

Include optional simple HTTP and FTP servers for demos where agents need to download or exfiltrate test files. Also include Jupyter notebook support for experimenting inside a VM and exposing notebooks on a chosen network interface.

Add clear setup instructions, a safety warning to only run this in a virtual or test environment, and basic code quality guidance. Look up current docs online if you need to.