Abdullah-Al-Raju/hexstrike-ai ? reverse-engineered prompt

Build me a Python based MCP server that lets an AI assistant connect to security tools and run automated pentesting style workflows from one place. I want it to work with MCP compatible clients like Claude Desktop, Cursor, and Copilot, and I want a simple way to start the server, check health, and send a target for analysis.

Please include the core pieces I’d expect from a real working version, like a server entry point, an MCP wrapper, basic API routes for target analysis, and support for running external security tools in a controlled way. It should feel like an autonomous security automation platform, with smart tool selection, progress updates, and some kind of reporting or visual feedback if that makes sense from the existing files.

Also make the setup easy to follow, with clear install steps, client integration examples, and sensible defaults. If you need to look up current docs online for MCP or any library behavior, go ahead and do that.