wahyunuriman999/Ruhci-Claude-Engine ? reverse-engineered prompt

Build me a Python project called Ruhci Claude Engine that helps an AI coding agent pick the right files and the right context instead of reading a whole repo.

I want a working command line tool that can scan a codebase, analyze files with static parsing, rank what matters most, and return a clean context package for an AI task. It should be able to keep basic memory across runs, plan simple steps, route tasks, and run from a central orchestrator. I also want a few demo commands like version, status, and run, plus a small example that shows it analyzing a module and printing the best files to read first.

Make the core pipeline actually work end to end, and keep the more advanced subsystems in place even if they are still early. Use Python, add tests for the main pipeline, and if you need current docs for anything, look them up online.