HubLensLLMsafishamsi/graphify
safishamsi

graphify

AI🌱 NEW PROJECT BOOST#AI Assistant#Knowledge Graph#Codebase Analysis#LLM#Developer Tools
View on GitHub
63

// summary

Graphify is an AI coding assistant skill that builds a comprehensive knowledge graph from your codebase, documentation, and multimedia files. It uses deterministic AST extraction for code and parallel subagents for conceptual analysis to map relationships without relying on embeddings. The resulting interactive graph and audit reports provide developers with deep architectural insights and efficient navigation across complex projects.

// technical analysis

Graphify is an AI coding assistant skill designed to transform diverse project assets—including code, documentation, media, and configuration files—into a structured, queryable knowledge graph. By utilizing a multi-pass approach that combines deterministic AST extraction for code with LLM-driven semantic analysis for unstructured data, it enables AI agents to navigate codebases using structural relationships rather than simple keyword searches. This design significantly reduces token consumption and improves architectural understanding by providing agents with a persistent, topological map of the project, effectively solving the problem of context fragmentation in large or complex repositories.

// key highlights

01
Builds a persistent knowledge graph from code, PDFs, images, and audio/video files to provide a unified view of project concepts.
02
Uses deterministic AST extraction for 25 programming languages to map code structure without relying on LLM inference.
03
Employs Leiden community detection on graph topology to identify clusters and 'god nodes' without requiring separate vector embeddings.
04
Integrates with major AI coding assistants like Claude Code, Cursor, and Aider to provide 'always-on' architectural context via pre-tool hooks.
05
Supports advanced graph operations including path tracing, semantic queries, and MCP server hosting for structured data access.
06
Maintains a SHA256-based cache to ensure efficient incremental updates, processing only changed files during subsequent runs.

// use cases

01
Automated codebase architectural mapping and dependency visualization
02
Multimodal knowledge extraction from code, PDFs, videos, and technical diagrams
03
Always-on AI assistant integration for context-aware code navigation and querying

// getting started

Install the tool using 'uv tool install graphifyy' or 'pipx install graphifyy'. Once installed, run 'graphify install' to configure your preferred AI coding assistant, then navigate to your project directory and execute '/graphify .' to generate the initial knowledge graph and report.