Skip to content

Graph Traversal

Graph traversal tools help you navigate and analyze the connections between documents in your Zettelkasten.

Overview

Your Zettelkasten forms a knowledge graph where: - Nodes = Documents - Edges = Wikilinks between documents

Graph traversal tools help you explore this structure.

Available Tools

Purpose: Find all wikilinks in a document

Example: 

You: What links are in "Getting Things Done"?

AI: Using tool: extract_wikilinks
Document: Getting Things Done.md

Found 5 wikilinks:
1. [[Next Actions]] (line 12)
2. [[Context Lists]] (line 24)
3. [[Weekly Review]] (line 45)
4. [[Project Planning]] (line 67)
5. [[Capture System]] (line 89)

Purpose: Discover what documents link TO a target document

This answers: "What references this document?"

Example: 

You: What documents link to "Getting Things Done"?

AI: Using tool: find_backlinks
Target: Getting Things Done.md

Found 5 documents with backlinks:
1. Daily Workflow.md (2 references)
2. Productivity System.md (1 reference)
3. Project Management.md (3 references)
4. Weekly Planning.md (1 reference)
5. Task Management.md (2 references)

Purpose: Discover what documents a source document links TO

This answers: "What does this document reference?"

Example: 

You: What does "Getting Things Done" link to?

AI: Using tool: find_forward_links
Source: Getting Things Done.md

Found 5 forward links:
1. Next Actions.md
2. Context Lists.md
3. Weekly Review.md
4. Project Planning.md
5. Capture System.md

Purpose: Find connection paths between documents

This answers: "How are these concepts connected?"

Example: 

You: How are "Deep Work" and "Flow State" connected?

AI: Using tool: link_path_finding
Source: Deep Work.md
Target: Flow State.md

Found 2 connection paths:

Path 1 (2 hops):
Deep Work → Focus Techniques → Flow State

Path 2 (2 hops):
Deep Work → Productivity Methods → Flow State

Purpose: Analyze connectivity and identify hub documents

Example: 

You: What are my most connected documents?

AI: Using tool: link_metrics

Hub Documents (by total links):
1. Productivity System.md (45 links)
   - Incoming: 28
   - Outgoing: 17

2. Getting Things Done.md (32 links)
   - Incoming: 18
   - Outgoing: 14

3. Note-Taking Methods.md (29 links)
   - Incoming: 21
   - Outgoing: 8

Use Cases

Understanding Structure

Find hub documents: 

You: What are my most important notes?

AI: [Uses link_metrics to identify highly connected documents]

Discovering Connections

Find relationships: 

You: How do my ideas about productivity relate to creativity?

AI: [Uses link_path_finding to discover connections]

Explore related content: 

You: Show me what's related to this document

AI: [Uses find_backlinks and find_forward_links]

Quality Assessment

Check orphaned documents: 

You: What documents have no links?

AI: [Uses link_metrics to identify isolated documents]

Graph Concepts

Definition: Documents that link TO a target document

Use: Understanding what references a concept

Example: 

Getting Things Done.md ← Daily Workflow.md
Getting Things Done.md ← Project Planning.md

Definition: Documents that a source document links TO

Use: Understanding what a concept references

Example: 

Getting Things Done.md → Next Actions.md
Getting Things Done.md → Context Lists.md

Definition: When two documents link to each other

Significance: Strong conceptual relationship

Hub Documents

Definition: Documents with many links (in + out)

Characteristics: - Central to your knowledge graph - Connect many concepts - Often important reference documents

Orphaned Documents

Definition: Documents with no links

Considerations: - May be new/incomplete - May be independent topics - May need integration

Best Practices

Good wikilinks: 

The [[Getting Things Done]] system helps with [[Task Management]].

Consistent naming: - Use the same title format - Match actual document names - Be consistent with capitalization

Building Connections

  1. Link related concepts - Connect similar ideas
  2. Create hub documents - Build index/MOC pages
  3. Bidirectional linking - Link back from referenced documents
  4. Semantic links - Link based on meaning, not just keywords

Analysis Workflow

1. Identify topic of interest
2. Find backlinks (what references it)
3. Find forward links (what it references)
4. Find paths to other concepts
5. Identify connection patterns
6. Build understanding

Visualizing Connections

While zk-chat doesn't provide graph visualization, the tools help you:

  • Understand structure - Through metrics and paths
  • Navigate effectively - Following link chains
  • Discover patterns - Identifying hubs and clusters

Example Queries

Discovery

"What are my most connected notes?"
"Show me hub documents"
"What links to this concept?"

"How do I get from [A] to [B]?"
"What connects these two ideas?"
"Show me the path between these documents"

Analysis

"What documents have no links?"
"What's the structure of my productivity notes?"
"How interconnected is my vault?"

Technical Details

Supported formats: 

[[Document Name]]
[[Document Name|Display Text]]
[[folder/Document Name]]

Path Finding Algorithm

  • Uses breadth-first search
  • Finds shortest paths
  • Returns multiple paths if available
  • Maximum depth configurable

Performance

  • Fast for small-medium vaults (<1000 docs)
  • Scales well with vault size
  • Results cached where appropriate

See Also