wtf is graphify?

graphify is a tool that plugs into AI coding assistants like Claude Code, Cursor, Codex, and Gemini CLI, and turns an entire project, including code, documents, PDFs, images, and video, into a knowledge graph you can ask questions against instead of manually searching through files. Typing a single slash command in your assistant maps everything into a graph made of concepts and the connections between them. Code itself is parsed on your own machine using tree-sitter, a code parsing library, so mapping code does not call any AI model and nothing about your codebase leaves your computer. Documents, PDFs, images, and video go through a semantic pass using your assistant's own model or an API key you configure, since those file types need language understanding rather than just structural parsing. Every connection in the graph is labeled as either explicitly found in the source or inferred by graphify's own resolution process, so you can always tell which is which. Running it produces three files: an interactive HTML page you can open in a browser to click through nodes and search, a short written report highlighting key concepts and interesting connections, and a raw graph file you can query later without re-reading your project. From your AI assistant you can then ask it to explain a specific concept, find the shortest path between two things in your codebase, or answer a plain language question, and get back an answer built from that graph rather than a fresh search each time. The project reports comparisons against other memory and retrieval systems on public benchmarks, claiming stronger recall while using no AI credits to build the graph itself, since the code mapping step runs entirely locally.