graphify

Builds persistent, cross-document knowledge graphs with auditable edge labeling from a folder of files.

Updated Apr 26, 2026

Search Tags:#graph #knowledge_graph #audit_trail #data_analysis #community_detection #cross_repo

Ready to run

npx skills add https://github.com/GongXLiang/plot-study --skill graphify-gongxliang

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: graphify
Source: https://github.com/GongXLiang/plot-study/tree/main/.claude/skills/graphify
Command: npx skills add https://github.com/GongXLiang/plot-study --skill graphify-gongxliang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turn any folder of files into a navigable knowledge graph with community detection, an honest audit trail, and three outputs: interactive HTML, GraphRAG-ready JSON, and a plain-language GRAPH_REPORT.md.

Core Features & Use Cases

Persistent graph with cross-session storage in graphify-out, preserving relationships across runs.
Honest audit trail for edges, tagging them as EXTRACTED, INFERRED, or AMBIGUOUS to separate discovered vs inferred knowledge.
Cross-document discovery and merging, enabling cross-repo graphs and agent-driven semantic extraction.
Outputs include an interactive HTML visualization, GraphRAG-ready JSON, and a human-readable GRAPH_REPORT.md.

Quick Start

Run graphify on a local folder to build a navigable knowledge graph and output the results.

Frequently Asked Questions about graphify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I build a knowledge graph from a folder of mixed code and documents?▼

To build a knowledge graph from a folder of mixed files, you process the directory to parse code, docs, papers, and images, clustering concepts into communities. This creates a persistent graph with cross-document relationships stored locally.

Can I merge knowledge graphs across multiple repository folders?▼

Yes, you can merge knowledge graphs across multiple repository folders using cross-document discovery and merging. This supports cross-repo graphs and optional agent-driven semantic extraction to connect concepts across distinct codebases.

What output formats are generated when converting files into a navigable knowledge graph?▼

Converting files into a navigable knowledge graph generates three outputs: an interactive HTML visualization, GraphRAG-ready JSON for retrieval pipelines, and a plain-language GRAPH_REPORT.md for human-readable summaries.

Does the knowledge graph persist across multiple analysis sessions?▼

Yes, the knowledge graph persists across multiple analysis sessions through persistent cross-session storage in a graphify-out directory. This preserves all discovered and inferred relationships across subsequent runs.

What is community detection and how does it organize knowledge graph concepts?▼

Community detection in a knowledge graph organizes related concepts into clustered groups based on their connections. This groups related code, documents, and papers together, making the navigable graph easier to analyze.