shubhrad1/LLMDB — reverse-engineered prompt

Reverse engineered prompt

GitHub

Build me a simple Python command line app called LLMDB that acts like a natural language helper for databases. I want to be able to run it, chat with it about database concepts in plain English, and also ask it to create and save database configuration files for me. It should feel like an interactive assistant, keep the conversation going until I type end, and save any generated configs into a temp folder.

Please wire it up so I can add an LLM API key in a .env file and start it with one command. Keep the code organized enough that I can later expand it into Docker based database setup and natural language query execution, but for now make sure the working parts are the chat experience and the config builder. If something in the repo is inconsistent, like file names or missing folders, just fix it so the project runs cleanly. Add a short README update with setup and usage, and look up current docs online if you need to.

Want more depth? Deep Reverse