Graphify: Transform Your Codebase into a Queryable Knowledge Graph

Introduction

Graphify is an advanced AI coding assistant skill designed to revolutionize how developers interact with their codebases. It takes any folder of code, SQL schemas, R scripts, shell scripts, documents, papers, images, or videos and transforms them into a queryable knowledge graph. This means you can have your application code, database schema, and infrastructure all represented in one interconnected graph, making complex projects easier to understand and navigate. Graphify integrates with a wide range of AI coding assistants, including Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and many more.

Installation

Getting started with Graphify is straightforward.

Prerequisites

Ensure you have Python 3.10+ installed. For package management, uv is recommended, or pipx as an alternative.

Python: 3.10+
uv (recommended): curl -LsSf https://astral.sh/uv/install.sh | sh
pipx (alternative): pip install pipx

Step 1: Install the package

The official PyPI package is graphifyy (note the double-y), though the CLI command remains graphify.

# Recommended (uv puts graphify on PATH automatically):
uv tool install graphifyy

# Alternatives:
pipx install graphifyy
pip install graphifyy

Step 2: Register the skill with your AI assistant

After installation, register Graphify with your preferred AI assistant.

graphify install

Now, open your AI assistant and type /graphify . to start mapping your project. For platform-specific installations, refer to the official documentation.

Examples

Graphify simplifies project exploration with intuitive commands.

Build a Knowledge Graph

To generate a knowledge graph for your current directory, simply run:

/graphify .

This command will produce three key files in the graphify-out/ directory:

graph.html: An interactive visualization of your graph, viewable in any browser.
GRAPH_REPORT.md: A summary of highlights, key concepts, surprising connections, and suggested questions.
graph.json: The full graph data, which you can query directly.

Querying the Graph

Instead of grepping through files, ask your graph questions:

/graphify query "what connects auth to the database?"

Export Call-Flow Diagrams

Generate readable architecture pages with Mermaid call-flow diagrams:

graphify export callflow-html

Add External Content

Integrate external resources like research papers or videos into your graph:

/graphify add https://arxiv.org/abs/1706.03762
/graphify add <youtube-url>

Why Use It

Graphify transforms how you understand and interact with complex codebases.

Deep Codebase Understanding

It provides a high-level overview of your project's architecture, revealing "God nodes" (most-connected concepts) and "surprising connections" between different modules or files. This helps in quickly grasping the core logic and interdependencies.

AI-Powered Insights

The generated GRAPH_REPORT.md offers AI-curated insights, including suggested questions the graph is uniquely positioned to answer, and explanations for code sections (from inline comments and docstrings). Every inferred relationship is tagged with EXTRACTED, INFERRED, or AMBIGUOUS, giving you confidence in the data.

Comprehensive File Support

Graphify handles a wide array of file types, ensuring a holistic view of your project:

Code: 31 languages including Python, TypeScript, Go, Java, C++, SQL, and more.
Docs: Markdown, HTML, TXT, RST, YAML.
Office: .docx, .xlsx (with optional extras).
PDFs: .pdf.
Images: .png, .jpg, .webp, .gif.
Video / Audio: .mp4, .mov, .mp3, .wav, and YouTube URLs (with optional extras).

Seamless AI Assistant Integration

By registering Graphify as a skill, your AI assistant can leverage the knowledge graph for codebase questions, preferring scoped queries over raw file searches, significantly enhancing its utility and accuracy.

Graphify: Transform Your Codebase into a Queryable Knowledge Graph

Summary

Repository Information

Topics

Use at your own risk