Hi everyone —
I’m still pretty new to the world of local LLMs and building prompt workflows, but one thing that kept frustrating me was losing memory every time a session ended.
I was constantly copying old prompts, trying to remember what I told the model, and rebuilding my system logic from scratch.
So I built a tool to help fix that.
It’s called the White Save Suite, and it gives you a way to save your memory, context, tone, or structure between sessions — even if your model has no built-in memory or you’re just using a chat box.
What it does:
-
Gives you 3 “save slots” for storing summaries, system prompts, or logic
-
Lets you use a simple flagging system (
= save,
= confirmed, etc.)
-
Helps you organize your sessions with automatic summaries and backups
-
Works in ChatGPT, LM Studio, Ollama, or any model with a text interface
You don’t need to install anything. It’s all copy-paste and human-readable.
Download or check it out here:
I’d love feedback — or just to hear how other people are handling memory in their own setups. Especially if you’re like me and tired of starting from scratch every time you open a new chat.
Hope it helps!
—Kevin