ailabs-london-2025

Gemini CLI Demo

This project serves as a demonstration of the Gemini CLI's capabilities, showcasing its ability to:

• Code Generation: Develop both backend API services and frontend user interfaces.
• Multi-modal Understanding: Utilize an image as a reference for designing and implementing the frontend UI.

Project Overview

The application consists of a simple FastAPI backend that provides a /greetings endpoint. This endpoint returns a personalized greeting if a name parameter is provided, or a default "hello world" message otherwise.

The frontend is a basic HTML, CSS, and JavaScript application that interacts with the backend. It allows users to input a name and retrieve a greeting, with its design inspired by a provided image reference.

How to Run

This project is designed to be built and run using the Gemini CLI. Follow these steps:

1. Start the Gemini CLI.
2. Run the Backend Prompt: Provide the PROMPT_BACKEND.md file to the Gemini CLI to create the API server.
3. Test the Backend Server: Once the server is running (you will be instructed to run uv run uvicorn main:app), you can test it by accessing the following URLs in your browser or with a tool like curl:
   • http://localhost:8000/greetings
   • http://localhost:8000/greetings?name=foo
4. Run the Frontend Prompt: Provide the PROMPT_FRONTEND.md file to the Gemini CLI. Please make sure to check image.png as instructed in the prompt, as it serves as the design reference for the UI.
5. Access the UI: After the frontend is set up and the server is running, open your web browser and navigate to http://127.0.0.1:8000 to interact with the application.