0x6b6679/worm-gpt ? reverse-engineered prompt

Build me a simple Python CLI for experimenting with LLM workflows and stress testing them in a messy, chaotic way.

I want it to feel like a small command line tool where I can load prompts from JSON files, run a few different test scenarios, and see the responses clearly in the terminal. It should be easy to add new test cases, save results, and compare runs over time. If there are any missing details from the old project, please keep it practical and make smart choices, and look up current docs online if you need to.

Make it clean enough that I can run it locally without much setup, and include a basic README with how to use it. Since this project was moved and the old repo is archived, just rebuild the useful parts in a straightforward way rather than trying to overcomplicate it.