How does this compare with other AI/LLM Note-takers and PKMs?

[BradKML]

There are some considerations when using note takers:

1. Whether it is based on folder hierarchies within notebooks, or projects with bundle-of-notes
2. Whether it is desktop-first, browser-first (seems not to be ideal), or mobile-first (needs to be lightweight in features)
3. Whether it has version control for fast rollback (Git adds complexity), or procedures to "hide" deprecated notes (easy blocks)
4. Whether it is article-based (Notion-like convention) or block-based (faster connecting with Obsidian-like structure)
5. Whether the app is AI-native (Notion and SiYuan), or requires plugins to add functionality (Obsidian and LogSeq)
6. Whether the AI involved can use tools to modify files agentically or not Enhance SiYuan with Obsidian plugin features and consider renaming to "SiYuan Copilot" jaar23/nb-assistant#6
7. (maybe there are other issues I might not know of yet, pls discuss)

For those LLM-enabled Note-taking tools

• https://github.com/learn-anything/learn-anything
• https://github.com/blinkospace/blinko
• https://github.com/satellitecomponent/Neurite
• https://github.com/signerlabs/Klee

For explicit "PKMs" (personal knowledge managers)

• Top FOSS PKM with DB refactoring https://github.com/logseq/logseq
• Second place FOSS PKM https://github.com/siyuan-note/siyuan
• VSCode with https://github.com/foambubble/foam and https://github.com/dendronhq/dendron
• AI-centric PKM https://github.com/reorproject/reor
• Org Mode https://github.com/org-roam/org-roam
• Wiki-ware derivative https://github.com/tiddly-gittly/TidGi-Desktop
• https://github.com/TuanManhCao/digital-garden
• https://github.com/lumen-notes/lumen

[brandonmarkgaia-hub]

Hey, I’ve been experimenting with a slightly different angle on AI + PKM: instead of just a note-taker, I built a public “codex” / digital garden that is driven by GitHub + Actions and then read by AI tools as a kind of symbolic world.

It treats GitHub Issues as “echoes”, markdown books as a long-form codex, and GitHub Actions as the autonomic system (indexing, signatures, status JSON, etc.). The idea is that any LLM can sit on top of it and treat it as a living world / knowledge garden rather than just a vault of notes.

I’m curious how this fits into your mental map of AI PKMs. Does it feel closer to a digital garden, an agentic note-taker, or something else?