0x4m4/hexstrike-ai ? reverse-engineered prompt

I want to build an advanced AI cybersecurity platform. The core idea is a Python server that acts as a bridge, connecting my AI assistant like Claude or GPT to a whole suite of real-world penetration testing tools. I want to be able to give a command to my AI, and it should use the server to autonomously run security tests on a target.

The system needs to be smart, capable of analyzing a target and then selecting the best tools for the job from a library of over 150 different ones—covering things like network recon, web app testing, and cloud security. It should have specialized AI agents for tasks like bug bounty hunting or CTF solving. The final output should be a clear visual report, maybe with "vulnerability cards," that's easy to understand. Make sure it's designed to integrate easily with clients like Cursor and VS Code.