Turn codebases, docs, and PDFs into a queryable knowledge graph for AI assistants.
Local tree-sitter AST parsing replaces grep with a real graph for your AI coding assistant.
Graphify is a local tool that parses your code, docs, and PDFs into a real knowledge graph rather than a vector store. You type /graphify in assistants like Claude Code or Cursor to query code paths instead of grepping files.
It uses tree-sitter locally to build deterministic AST graphs of code across 40 languages without sending code to an LLM. It generates three output files including an interactive HTML graph, a markdown report, and a JSON file for querying.
It suits developers and executive assistants managing complex codebases who want to cut token spend and trace code connections. It does not suit teams unwilling to run a Python CLI or install assistant skills.
Parses code locally using tree-sitter without sending your source files to an LLM.
Every connection is tagged as explicitly extracted from source or inferred by resolution.
Produces a clickable browser-based HTML graph alongside markdown summaries and raw JSON.
Integrates via a single command into Claude Code, Cursor, Codex, Gemini CLI, and GitHub Copilot.
Free and open source under the Apache-2.0 license, with optional paid Graphify Cloud tiers for continuous monorepo monitoring and team integrations.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
No, it builds a real graph you traverse rather than using embeddings or vector stores.
It supports Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, and over 20 others.
Code is parsed locally on your machine using tree-sitter AST, so nothing leaves your machine.
See Zekai first in Google