techtony2018/memory-stargraph — explained in plain English
Analysis updated 2026-05-18
Visually explore a gbrain knowledge base as a clickable star-cloud graph.
Ask AI questions grounded in the content and relationships of a selected note.
Edit tags, relationships, and attachments directly from the graph interface.
Run an automatic Autopilot tour through important nodes for a demo or review.
| techtony2018/memory-stargraph | aim-uofa/reasonmatch | airbone42/360-data-athlete | |
|---|---|---|---|
| Stars | 12 | 12 | 12 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | hard | hard |
| Complexity | 3/5 | 5/5 | 4/5 |
| Audience | developer | researcher | general |
Figures from each repo's GitHub metadata at analysis time.
Requires the gbrain knowledge system to already be installed and configured before use.
Memory Stargraph is a local web tool for exploring a knowledge base, built on a system called gbrain, as an interactive map of connected topics that looks like a cluster of stars. Instead of scrolling through a list of notes, you see every entity in your knowledge base as a node you can click, drag into view, and drill down from, with related items loading automatically as you explore. It is built with plain Python and basic web technology, so it runs without installing any extra web development tools. The main feature is called Ask Yoda, a chat window tied to whichever node you have selected. When you ask a question, the tool gathers the content of that page, its nearby relationships, related search results, and other connected notes, then sends all of that context to an AI system to produce an answer. If a question refers to another entity by name, it will specifically look up that entity and follow its connections to find a more precise answer. There is a depth setting that controls how much surrounding context gets pulled in, trading speed for thoroughness. A diagnostics panel lets you see how a given answer was put together, without exposing anything private. Beyond viewing, you can edit your knowledge base directly from the interface: add tags, timeline events, attachments, and relationships between entities, or hide nodes you do not want cluttering the map. Files and other media can be attached to entries alongside their text and previewed in place. Search understands exact matches to entries as well as broader indexed lookups, and it correctly handles non-English text. An automatic tour mode called Autopilot can walk through a generated or custom list of important nodes on its own, which is useful for demonstrations or for rediscovering parts of a knowledge base you have not looked at in a while. The tool also caches data locally in the browser so it stays responsive when the underlying knowledge system or network connection is slow, and it includes performance tuning so that even large, dense graphs remain smooth to navigate. Using Memory Stargraph requires the gbrain knowledge system to already be set up and working on your machine, since Memory Stargraph is a viewer and editor built on top of it rather than a standalone knowledge base of its own.
A local web app that visualizes a gbrain knowledge base as an interactive star map and lets you chat with AI about any selected node.
Mainly Python. The stack also includes Python, HTML, CSS.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.