skills.rest
singhaganesh
singhaganeshCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

graphify-windows

Convert files into a knowledge graph with community detection and HTML, JSON, and report outputs.

Updated Jul 7, 2026
Search Tags:#knowledge graph#obsidian#codebase analysis#neo4j#graphrag#community detection#ast extraction
One-click install
npx skills add https://github.com/singhaganesh/Urban-assist --skill graphify-windows-singhaganesh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: graphify-windows
Source: https://github.com/singhaganesh/Urban-assist/tree/main/.cursor/skills/graphify
Command: npx skills add https://github.com/singhaganesh/Urban-assist --skill graphify-windows-singhaganesh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires graphifyy.

What problem does it solve? Understanding a large codebase, research corpus, or mixed folder of documents requires reading everything manually, and insights disappear between sessions. This Skill turns any folder into a persistent, queryable knowledge graph with an honest audit trail showing which relationships were extracted versus inferred. ## Core Features & Use Cases - Knowledge Graph Extraction: Combines deterministic AST parsing for code with parallel AI subagents for docs, papers, and images, tagging every edge as EXTRACTED, INFERRED, or AMBIGUOUS with confidence scores. - Community Detection & Analysis: Clusters the graph into labeled communities, identifies god nodes and surprising cross-document connections, and generates a plain-language GRAPH_REPORT.md. - Multiple Output Formats: Produces interactive HTML visualization, GraphRAG-ready JSON, SVG, GraphML, Neo4j Cypher exports, Obsidian vaults, and an agent-crawlable wiki. - Query & Maintenance: Supports BFS/DFS natural-language queries, shortest-path lookups, incremental updates, file watching, and an MCP server for agent access. - Use Case: Point it at an unfamiliar repository to understand its architecture before touching code, or drop papers, tweets, and notes into a folder and query the combined concept graph weeks later without re-reading anything. ## Quick Start Ask the AI to run /graphify on the current directory to build a knowledge graph and open the generated report.

Frequently Asked Questions about graphify-windows

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

FAQPage Schema
How do I build a knowledge graph from a codebase?▼

Run /graphify on the project directory. It detects supported files, extracts code structure via AST parsing and semantic relationships via parallel subagents, then clusters the graph and outputs HTML, JSON, and a GRAPH_REPORT.md summarizing communities and key nodes.

What file types can be turned into a knowledge graph?▼

Code files, Markdown and text documents, PDF papers, images, and video or audio files are supported. Video and audio are first transcribed with Whisper, then treated as documents during extraction.

How do I update a knowledge graph when files change?▼

Use the --update flag for incremental rebuilds, which re-extracts only new or changed files using the semantic cache. The --watch flag monitors a folder and rebuilds automatically on code changes without needing an LLM.

Can I export a knowledge graph to Neo4j or Obsidian?▼

Yes. The --neo4j flag generates a Cypher script and --neo4j-push writes directly to a Neo4j instance. The --obsidian flag writes a vault of linked notes, and --graphml or --svg export to Gephi, yEd, Notion, or GitHub.

Why does graph extraction produce an empty graph?▼

An empty graph means extraction produced no nodes, usually because all files were skipped, the corpus contains only unsupported binaries, or extraction failed. Check the detection summary for skipped files and rerun on a folder with supported content.

What are the limitations of automated knowledge graph extraction?▼

Semantic extraction relies on LLM inference, so edges are tagged EXTRACTED, INFERRED, or AMBIGUOUS with confidence scores to distinguish found relationships from guesses. Very large corpora over 200 files or 2 million words require scoping to a subfolder first.

Related Skills

Explore More Software Engineering Skills→
mayconspirlandellimayconspirlandelli

mira-references

Organizes per-theme reference folders containing PDFs, images, and notes for slide decks.

Community
Basic
ulebuleulebule

graphify

Converts folders of code, docs, papers, and media into a queryable knowledge graph with community detection.

Community
Advanced
rohanbhatia27rohanbhatia27
1

audit-prep-assistant

Prepare codebases for security review with static analysis and documentation generation.

Community
Advanced
plurigridplurigrid
60

cider-embedding

Generate semantic embeddings for Clojure code navigation via CIDER.

Official
Basic