Managing AI generated code
TODO
TODO Introduction
Objective
The objective is to answer the following questions. Given a code repository written by an AI like Lovable that has a bug i.e. that violates some invariant ; then should one try to fix the bug through Lovable or should one try to fix the code manually? Then, how to ensure non-regression?
TODO Lovable
Documentation
From lovable.dev/pricing :
> What tech stacks does Lovable know?
Lovable builds front-end using React, Tailwind & Vite, and can connect to OpenAPI backends. Support for data persistence and authentication is in alpha, using Supabase.
From docs.lovable.dev/introduction/welcome :
Lovable supports the full lifecycle of modern product development, from early exploration and prototyping to deployment and ongoing operation.
Each project represents a single application and produces a codebase that can be synced to GitHub and integrated into existing engineering workflows.
Lovable is designed to complement existing tools and processes:
- Describe what you want to build
- Review and iterate on the generated application
- Sync code to GitHub
- Deploy, operate, and govern according to your standards
Getting started
From docs.lovable.dev/introduction/getting-started :
Simply enter an initial prompt to kick things off, and Lovable will do the rest! This is the starting block for any project in Lovable, where you can bring your ideas to life instantly.
Prompt :
Create a TODO application.
Github transfer
After the transfer, here is the github repo :