Tana Supertags Guide — source summary
A third-party guide to tana and its “supertags,” chased as the final PKM-tools gap.
Capture in raw/tana-supertags-guide.md (Tana’s own docs redirected/404’d).
Key points
- Tana is an “AI-native workspace”: an outliner over a graph data model that turns notes into structured, queryable data.
- Supertags turn any node into a typed object with fields (e.g. #Task with due date/priority/project; #Meeting, #Person, #Project), with inheritance from base types.
- Everything is a node; nodes link via reference fields. Live Queries auto-update (e.g. “#Task, Priority::High, Due::next 7 days”).
- AI-native: supports GPT-5.1, Gemini 3 Pro, Claude Sonnet 4.5; AI uses your structured supertag data as context (your graph, not a blank doc); voice memos auto-tag and file themselves.
Why it’s the closer for cluster A
See tana — it fuses the two strands (roam-research networked nodes +
notion structured fields), its supertags are typed objects (a third independent
convergence on typed notes, with gbrain‘s schema packs and this wiki’s schema.org
@type), and it is AI-native — the LLM reasoning over your structured personal graph,
which is the llm-wiki/gbrain thesis arriving inside a mainstream PKM tool.