README
the first lines# Membase use cases
Nineteen things that long-term memory makes possible, each built as a small app you can run.
Every app calls the real Membase API and a real model; nothing is mocked. Each one has a video
of about a minute.
## The two bases
Every demo is written once against the same SDK (`membase.Membase`) and runs on either base:
| Base | What it is | Use it for |
| hosted | The Membase service, with a developer key from the app (Connect › Developer keys) | One person and their AIs. This is the product itself. |
| local | The same API, answered in process by the local engine (`unibaseio-membase[local]`), one directory per end user | Apps that serve many people, such as companions and games, where each person's memory lives in its own directory |
Each demo has a default base. Set `MEMBASE_DEMO_MODE=local|hosted` to switch. Agent memory
(cases and skills learned from agent runs) exists only on the local engine today.
## Setup
Ask a scout
a deep read
GeminiGoogle . Gemini 3.8 Flash for the deep read
unibaseio is in the AI agents camp, which Gemini reads.
10 of 10 public reads left today. Holders get 3 a day.
No read kept yet.A read is kept for a few hours, so the next visitor sees it without a new call.
Scout calls and heat are information, not advice. They describe public GitHub activity and are never a signal to trade.
Recent commits
default branch- 1dAdd 18 Video memory and 19 Photo memory . ibitnoah
- 14dEvery video opens with the picture explainer . ibitnoah
- 14dVideo openings explain in pictures . ibitnoah
- 14dVideos: attachments uploaded to the public repository . ibitnoah
- 14dMembase use cases . ibitnoah
Releases
newest firstNo releases.This repo has not published a release on GitHub.