AashishSK333/medical_summary_agent — reverse-engineered prompt
Reverse engineered prompt
Build me a simple educational Python demo that shows how AI agents can work together to create a medical research summary.
I want it to run from the command line. The user should be able to give a health topic, then one agent acts like a medical researcher and searches demo PubMed style results, and another agent turns that research into a clear markdown report. Use CrewAI for the agent teamwork and AWS Bedrock with Amazon Nova Pro for the language model. Include basic memory so the demo can show how context is kept between steps.
Please keep it easy to understand for learners. Add clean setup instructions, an example env file for AWS settings, simple config files for the agents and tasks, and a sample output report. Include a quiet mode and a default topic if the user does not provide one. Add basic tests for the configuration and make sure the project can be installed and run with uv. Look up current docs online if you need to.
Want more depth? Deep Reverse