Map an unfamiliar codebase into a browsable, clickable knowledge graph.
Ask your AI coding assistant to explain a concept and trace its connections instead of grepping files.
Find the shortest path between two ideas or components in a large project.
Fold documentation, PDFs, and design notes into the same graph as the code.
| graphify-labs/graphify | github/spec-kit | hacksider/deep-live-cam | |
|---|---|---|---|
| Stars | 93,665 | 92,895 | 92,686 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | hard |
| Complexity | 3/5 | 3/5 | 3/5 |
| Audience | developer | developer | vibe coder |
Figures from each repo's GitHub metadata at analysis time.
Graphify is a tool that turns an entire codebase, along with its documentation, PDFs, configuration files, and even images or video, into a knowledge graph, a map of concepts and how they connect to each other. Instead of searching through files with a text search, you can ask questions and trace connections directly against this map. It works as a skill you add to an AI coding assistant such as Claude Code, Cursor, Codex, or Gemini CLI. You install a small command line tool, register it with your assistant, then type a single command inside your project. The tool reads your code using a parsing technique that understands the actual structure of each programming language, rather than relying on an AI model to guess, so this part runs entirely on your own computer with nothing sent elsewhere. Documents, PDFs, images, and video use your assistant's AI model, or a separately configured one, for a more meaning based pass. Once built, the graph is saved as three files: an interactive page you can open in a browser and click through, a written report summarizing the key concepts and interesting connections, and a raw data file you can query later without re-reading your project. Every connection in the graph is labeled as either directly found in the source code or worked out through inference, so you always know how certain a given link is. On top of the graph, you get commands to explain a single concept and its connections, find the shortest path between two ideas in your codebase, or ask a plain language question and get back a focused piece of the graph relevant to it. The project also groups related code into clusters automatically, highlights the most connected, central concepts, and pulls in explanatory code comments and design documents as their own linked entries. The README also compares Graphify's approach against other memory and retrieval systems on a couple of published benchmarks, generally reporting stronger or comparable results while using no AI credits to build the graph itself. The full README is longer than what was shown.
A tool that maps your entire codebase, docs, and files into a queryable knowledge graph for AI coding assistants.
Mainly Python. The stack also includes Python, tree-sitter, Claude Code.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.