G
Graphify
graphify.homes
Open-source knowledge graph skill

Knowledge Graphs for AI Coding Assistants

Graphify turns code, docs, PDFs, screenshots, diagrams, and transcripts into one queryable graph so your assistant can answer architecture questions with structure instead of guesswork.

Cloud quick start
Sign in → Upload ZIP → Explore graph (graphify.homes)
No install required for the hosted product. Optional local tooling is documented on the open-source repo.
Token compression
71.5x

Mixed corpus benchmark from the official project examples.

AST languages
23

Tree-sitter parsing across modern programming languages.

Outputs
3 core

graph.html, graph.json, and GRAPH_REPORT.md stored in your account.

License
MIT

Open-source, commercial-friendly, telemetry-free by default.

Live Graph Surface

Architecture at a glance

graph.html + graph.json
ingest.pyASTsemanticclusterreportdocsgod nodesurprisepath queryexplainpdfimagetranscripthtml export
Structural edges

Imports, calls, classes, docstrings, and rationale comments stay attached to the same graph.

Persistent memory

Query the graph weeks later without rereading the entire repository from scratch.

Exportable artifacts

Share a human-readable report with teammates or load the graph into your own tooling.

Core capabilities

Built for the messy reality of modern codebases

Keyword search can locate files. Graphify explains how those files, decisions, and artifacts connect.

Multimodal extraction

Read code, docs, PDFs, screenshots, diagrams, transcripts, and rationale in one pass.

Structure-first graph build

Combine AST edges, semantic links, and design notes into a queryable NetworkX graph.

Community clustering

Leiden clustering groups subsystems by topology, not by another embedding layer.

God nodes and surprises

Surface architectural gravity wells and unexpected cross-file connections worth investigating.

Web-first workflow

Upload a project ZIP and explore the graph in the browser—no local Python install required for the cloud product.

Secure-by-design

Strict input validation, bounded downloads, path containment, and escaped output labels.

Why it feels different

Query the why, not just the where

Graphify uses static analysis for hard structure, then enriches the corpus with semantic edges, rationale, and multimodal context. Instead of pushing every question through a fresh vector retrieval loop, it gives your assistant a durable graph of the system.

That means path explanations, community views, god nodes, and surprise edges are first-class outputs, not accidental artifacts of one lucky prompt.

01

Detect

Collect code, prose, visual references, and media into one corpus.

02

Extract

Use Tree-sitter and semantic extraction to produce nodes, edges, and rationale.

03

Build

Merge everything into a persistent graph instead of a disposable prompt context.

04

Cluster

Reveal subsystems with Leiden communities and highlight graph centers.

05

Explain

Answer graph queries, path lookups, and architectural why-questions.

06

Export

Ship graph.html, graph.json, GRAPH_REPORT.md, and incremental cache artifacts.

Comparison

Graph intelligence, not another pile of chunks

Capability
Graphify
Vector RAG
Plain search
Code structure awareness
Native AST edges, imports, calls, comments, and rationale.
Usually flattened into chunks before retrieval.
Keyword and symbol lookup only.
Multimodal corpus
Code, docs, PDFs, screenshots, diagrams, audio, and video.
Possible, but often requires custom ingestion pipelines.
Mostly text and code files.
Design intent retrieval
Captures why decisions were made and links them to implementation.
Depends on chunk quality and prompt luck.
Not modeled.
Persistent query layer
Queryable JSON graph survives beyond a single session.
Depends on external index infrastructure.
No graph memory.
Embeddings required
No. Community detection runs on graph topology.
Yes.
No, but no semantic graph either.
Hosted workflow

Build a graph without a local install

Upload a ZIP from the Build page. We extract it safely, run Graphify on our servers, and keep graph.html, graph.json, and GRAPH_REPORT.md in your account. You do not need Python or pip install in the browser workflow.

Sign in → Upload ZIP → Explore graph (graphify.homes)

Self-hosted CLI installs are only for contributors or IDE integrations—not required for this product.

graph.html
graph.json
GRAPH_REPORT.md
FAQ

Straight answers for technical buyers

How do I use Graphify on this site?+

Sign in with Google, upload a ZIP of your project from the Build page, and we run Graphify in the cloud. You get graph.html, graph.json, and GRAPH_REPORT.md saved to your account—no local install required.

Does Graphify send raw source code to a third-party model?+

The upstream Graphify pipeline may call your configured model for semantic passes; see the official project for details. This product runs builds on our servers with credentials you configure in deployment.

Can I still use assistants like Claude Code locally?+

Yes. The open-source Graphify project integrates with Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and other assistants. graphify.homes focuses on the hosted ZIP workflow.

What artifacts do I get back?+

A visual graph, a durable graph.json, a human-readable report, and a cache that makes repeat runs cheaper.

Graphify on graphify.homes

A focused keyword landing page built around official Graphify product facts: multimodal extraction, graph-native architecture, and assistant integration.