Built a tool to manage AI conversation memory


I have been frustrated by the lack of cross collaboration between AI assistants. I like to switch Gemini, Claude and ChatGPT, depending on task and model performance. The problem when they do not communicate with each other is I have to ALWAYS start from zero.

So, I built mindlock.io, which allows me to export my favorite conversations (simple HTML page save), import them into the app and distill them into memory documents, using a local LLM, and generate curated context that I can paste into any AI assistant, stored within the browser memory, fully local.

The main goal is building a knowledge base that's assistant-agnostic, instead of being locked to the small memory the AI assistants give us. And of course, better context = better answers.

AI power users, would you be willing to give this system a try and let me know what you would improve?

Leave a Reply