krishnaik06/Deep-agents-With-Langchain ? reverse-engineered prompt

Build me a simple demo app for experimenting with deep AI agents using LangChain.

I want a clean Streamlit interface where I can type a goal or question, send it to an agent, and watch it work through the task in a useful way. It should feel like a hands on playground for trying different agent behaviors, showing the prompt, the steps it takes, and the final answer in a readable layout. If it helps, include a small example workflow so I can test it right away.

Keep the project easy to run locally, with a main Python entry point and a Streamlit app. Make it polished enough that someone can explore the idea without setting up much else. Use current docs online if you need to, and make sure the code is organized so I can extend it later.