Great to hear! Excited to see it again.

Since we work with many clients, each with unique goals, I envision separate project spaces for each one. These spaces would house persistent instructions, client background information (including their communication style and overall mentality), and project-specific objectives. As you know Maple does have the chat history, which allows us to rename chats for organization. However, I’ve encountered issues with token limits – when chats become too long, the AI summarizes them and starts a new chat, sometimes losing context in the process.

Project spaces would address this by storing all relevant instructions and context within the project itself. We could then start new chats within each project space, ensuring the AI always has the necessary information. This would save time by eliminating the need to repeatedly provide context, mitigate the token limit issue, and improve overall organization compared to scrolling through chat histories. For example, when working on a report for Client Y, the AI would already understand their brand guidelines, project goals, and client communication styles with each new chat.