graphify

To build a knowledge graph from a folder of notes, you can extract wikilinks and frontmatter to create cross-document connections. This process applies canonicalization and clustering to generate a navigable graph with detectable document communities.

Yes, you can map codebases and research papers using a knowledge graph. The extraction process detects file relationships and infers hyperedges, organizing diverse file types like code, papers, and images into cross-document clusters for quick insights.

Wikilinks and frontmatter automate edge creation by providing regex-based extraction inputs for the document graph. The system uses these extracted file relationships alongside inferred hyperedges to connect multi-node patterns across the vault.

The best way to audit and deduplicate file relationships in a personal vault is through deterministic preflight and canonicalization. This process builds an audit trail that clusters documents and removes duplicate connections before generating a final report.

Building a navigable knowledge graph from files generates exportable outputs including graph.json, GRAPH_REPORT.md, and an HTML file. These outputs provide persistent, clusterable graph data and an annotated audit trail for quick repository insights.

No, you do not need to manually define document clusters when creating a file relationship graph. The system automatically applies clustering algorithms to the extracted and inferred edges, organizing the documents into communities automatically.