Your notes are a graph. Make the links mean something.

The plugin turns links into typed, signed, labeled edges with properties, queries them with openCypher and draws a graph that looks like what it means. Then the same vault becomes your AI's memory.

In the community plugin directory as Typed Graph. Desktop and mobile.

See it work

A Zettelkasten in the web demo, which runs the plugin's own engine and renderer: 24 notes, then a query that follows a highlight to the idea it became, and one that traces notes to books and writers.

Write edges like you write notes

Plain type:: [[Target]] lines from Graph Link Types keep working. Add a sign and properties when you need them.

knows:: [[Bob]] {since: 2020, label: "met at NeurIPS"}
works_at:: [[Acme]]
-distrusts:: [[Mallory]]

And now your AI can read it too

The vault is already the agent's memory. Point the sidecar at it for Cypher, vector search and cited GraphRAG over MCP, or run tg against the folder from Claude Code.

npm install -g @typedgraph/cli
tg search "who works with Acme"

Agents guide →

What you get

↗

Typed, signed edges

A - prefix for negative edges, properties in braces, stable or pinned ids.

◆

Schemas in plain notes

Types, allowed edges, templates and looks, declared in Markdown. They export as SHACL.

⌕

openCypher blocks

A graph-query block shows a graph or a table, live, as the vault changes.

◎

Graph view

A full-pane explorer that follows the active note and explains every style.