Hey everyone, I wanted to ask you all a question about your workflows. I have been using cursor and github copilot for a few years now and I have been moving into the context engineering side of it. I noticed a big issue has been grabbing the correct context associated to the feature or problem they are trying to solve. Most times these chats are ephemeral and the current version of “memories” don’t really do it justice from a recall standpoint. using RAG or grep on codebases doesn’t really solve the needle in a haystack problem. I created a simple tool that creates, retrieves, summarizes, and cleans up notes that are associated to specific files so you always have up to date notes from your AI Agents. I wanted to see if anyone else has run into this issue and has approached the problem in a different way or has advice on how to mold my open source solution. https://github.com/a24z-ai/a24z-Memory