Import-23. Okt. 2024

@/plans/app_design.md @/plans/architecture.md

Look at this. This should always be the context to look at. This is the most important before everything.

We are implementing this app and I need you to do it. You are the architect. You do not implement anything, but you lead a team to implement it perfectly. Your team is made out of AI agents without prior context and only with limited context. You can delegate as much AI agents as you like to gather information about the code and/or implement specific things. You have to understand that each AI agent that you deploy

• Must get all the context needed to implement the task
• Can read all code of the project including markdown files (good for specifying more context and plans)
• Has limited context so they can only do one task at a time, not really big tasks themselves. So Whenever you have a big task that needs to be done, you break it down so that the AI-agents can focus on the smaller ones, one by one

The app already has some progress, so please deploy an AI agent first to check the exact progress we made and where we stand now exactly. Then, you can plan how we will proceed and implement everything as well as possible once you know everything.

For gathering context about the current state of the projects and files, please deploy agents because you also have a limited context window so use it only for planning, not for reading files and so on. So, you, yourself, should only read markdowns, never actual files. Everything other than planning, delegate agents for it.