Get AI summaries of any video or article — Sign up free
ExcaliBrain is transforming how I use Obsidian.md thumbnail

ExcaliBrain is transforming how I use Obsidian.md

5 min read

Based on Zsolt's Visual Personal Knowledge Management's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

Excalibrain renders an Excalidraw-based, interactive graph of an Obsidian vault using Dataview as the data engine.

Briefing

Excalibrain is an interactive mind-map layer for Obsidian that turns a vault’s structure—folders, links, tags, and YAML front matter—into a navigable graph with explicit and inferred relationships. The payoff is practical: it replaces scattered navigation (file explorer, tag pane, and manual link chasing) with a single visual map that makes context appear instantly, including connections through “virtual notes” that don’t exist as files yet.

At its core, Excalibrain is built on two Obsidian plugins. Excalidraw provides the canvas engine for rendering and experimentation with automation, while Dataview acts as the data layer that extracts fields and link metadata from markdown. The result is a structured graph where the currently open document sits at the center, with children below, parents above, “friends” to the left (lateral connections), and siblings to the right. Colors and node types are customizable, and the view can be filtered—for example, hiding tag-driven siblings reduces visual noise and reveals only the structural children of a node.

Relationships in Excalibrain come in two flavors. Explicit relationships are defined via Dataview fields (typed as field names followed by a double colon in markdown). Inferred relationships are derived from link behavior: forward links imply parent/child direction, backlinks imply the reverse, and mutual links become lateral “friend” connections. Siblings are treated as the children of a node’s parents, which is why enabling tag/category fields can dramatically increase sibling counts.

Using Excalibrain changes how the vault is authored and navigated. The graph encourages more atomic notes—smaller, more linkable units—because large documents create cluttered views. To manage that, the workflow shifts toward embeds and tools like Node Refactor to split bigger pages into smaller parts. Even styling matters: lightweight CSS adjustments improve embedded transclusions so the graph stays readable.

A standout feature is support for virtual notes—links to items that haven’t been created as files. Excalibrain can still display their relationships, letting users navigate across the vault’s intended structure before the underlying markdown exists. That makes planning and refactoring feel less like a separate task and more like part of day-to-day navigation.

Excalibrain also functions as a control center. It can index and search nodes, exclude folders like templates to prevent clutter, cap the number of displayed children/friends (the presenter keeps it around 40), and apply per-node styling rules based on tags/categories. It supports workspace “pinning,” so the graph can either follow navigation in a specific pane or remain fixed, and it can open the current graph as an editable snapshot via a pencil button.

Installation requires Obsidian BRAT since Excalibrain isn’t in the community plugin store yet. Once installed, the graph becomes a live map of the vault—one that reflects both the information architecture and the link logic embedded in markdown.

Cornell Notes

Excalibrain turns an Obsidian vault into an interactive mind-map by rendering a graph on an Excalidraw canvas using Dataview as the data engine. The center node is the active document, with children below, parents above, friends to the left (lateral links), and siblings to the right; node appearance can be customized and filtered. Relationships come from two sources: explicit Dataview fields (typed in markdown) and inferred links (forward links, backlinks, and mutual links). The graph also supports virtual notes, so uncreated items can still appear with their inferred/explicit relationships, enabling navigation during planning. The workflow shifts toward atomic notes and strategic embeds to keep the graph readable, and Excalibrain can replace file explorer and tag-pane navigation with one integrated map.

How does Excalibrain decide what counts as a parent, child, friend, and sibling?

It uses two relationship types. Explicit relationships come from Dataview fields in markdown (field name + double colon). Inferred relationships come from link direction: if a file A links forward to file B, A becomes the parent and B the child; if there’s a backlink, the parent/child direction flips accordingly. Mutual links (two files pointing at each other) become lateral “friends.” Siblings are derived structurally: they’re the children of a node’s parents, which is why enabling tag/category-based fields can increase sibling counts.

What’s the practical difference between explicit and inferred relationships in the graph?

Explicit relationships are white in the presenter’s color scheme and are driven by Dataview fields like “working on” or “inspired by.” Inferred relationships are derived from how documents link and where they appear in the graph; for example, if a document links to “Obsidian Dataview,” Excalibrain infers Dataview as a child. This means the graph can reflect both authored semantics (fields) and structural link behavior (links/backlinks).

Why does Excalibrain push users toward atomic notes and strategic embeds?

Large documents tend to generate clutter because everything linked inside them becomes visible at once in the graph. Smaller, atomic notes reduce that density. The workflow also leans on embeds/transclusions so the graph can reference key nodes without exposing every internal detail. Tools like Node Refactor help split bigger pages into smaller parts, and CSS tweaks can improve how embedded content appears so transclusions don’t overwhelm the reading experience.

What are virtual notes, and why do they matter for navigation?

Virtual notes are links to items that don’t yet have a corresponding file. Excalibrain still renders them and preserves relationships (parents/friends) even before the markdown exists. That lets users click through connections across the vault’s intended structure—useful for planning tag/category systems or refactoring—without waiting to create every target note first.

How do Excalibrain settings control clutter and visual readability?

Settings include excluding specific folders from indexing (e.g., templates) and toggles to hide node types like virtual nodes, folders, tags, and attachments. There’s also a maximum node count per relationship type (the presenter keeps it around 40) to prevent graphs from becoming unreadable when notes have hundreds of links. Styling rules can be applied based on tags/categories, including node prefix, border color, and font family overrides.

What workspace features help integrate Excalibrain into daily Obsidian use?

Excalibrain can be pinned to a specific pane so navigation updates only within that workspace area, avoiding confusing cross-pane changes. A pencil button opens the current graph as an editable snapshot in a new pane; the snapshot can be saved under a new name, which is useful for demonstrations or adding commentary around a captured graph state.

Review Questions

  1. When would a relationship be inferred as a friend rather than a parent/child connection?
  2. How do Dataview fields enable explicit relationships, and what markdown syntax is used to define them?
  3. What strategies (note granularity, embeds, settings limits) help keep an Excalibrain graph from becoming cluttered?

Key Points

  1. 1

    Excalibrain renders an Excalidraw-based, interactive graph of an Obsidian vault using Dataview as the data engine.

  2. 2

    The active document sits at the center, with children below, parents above, friends to the left, and siblings to the right; node types and colors are customizable.

  3. 3

    Explicit relationships come from Dataview fields in markdown, while inferred relationships come from forward links, backlinks, and mutual links.

  4. 4

    Siblings are derived as the children of a node’s parents, so enabling tag/category fields can significantly change sibling counts.

  5. 5

    The graph workflow encourages atomic notes and strategic embeds/transclusions to avoid clutter from large documents.

  6. 6

    Virtual notes appear in the graph even before their files exist, preserving relationships for planning and navigation.

  7. 7

    Excalibrain isn’t in the community plugin store; installation uses Obsidian BRAT.

Highlights

Excalibrain can navigate through virtual notes—links to items that don’t yet exist as files—while still showing their relationships.
Forward links, backlinks, and mutual links map directly to parent/child and friend (lateral) relationships, letting the graph reflect link logic without extra manual setup.
A small settings cap (around 40 nodes per relationship type) and the ability to hide node categories are key to keeping the graph usable.
Pinning Excalibrain to a specific workspace pane prevents the graph from changing when navigation happens elsewhere.
Snapshots created via the pencil button turn the current graph view into an editable Excalidraw document for demos or annotations.

Topics

Mentioned