aianytime/multi-agents-system-from-scratch ? reverse-engineered prompt

Build me a simple Streamlit app in Python that shows how a multi agent AI system can work without using any agent frameworks.

I want three tools in the app. One that summarizes medical text, one that helps write a research article from a topic and optional outline, and one that removes private health information from medical data. Each tool should have a second pass that checks or improves the result, so the summary gets validated, the article gets refined, and the sanitized text is checked to make sure no sensitive info is left behind.

Use OpenAI for the AI responses, and make the app easy for beginners to follow. Add clear input boxes, buttons, and result sections, plus basic logging so I can see what happened if something goes wrong. If it helps, look up the current docs online before you wire it up. Also include an example env file for the API key and make sure it runs cleanly from a single app.py entry point.