Graphify is an open-source knowledge-graph tool with 130,000+ stars on GitHub that turns any codebase — plus its docs, SQL schemas, configs, and PDFs — into a queryable knowledge graph, so you can ask questions instead of grepping through files. Built by Graphify Labs (Y Combinator S26) and released under Apache 2.0, it installs in about thirty seconds with uv tool install graphifyy followed by graphify install, which registers a /graphify skill in your AI coding assistant.
Type /graphify in Claude Code, Cursor, Codex, or Gemini CLI and it maps your entire project into a graph you can traverse. Code is parsed locally with deterministic tree-sitter AST analysis — no LLM involved, nothing leaves your machine — while docs, PDFs, images, and video get a semantic pass from your assistant’s model or a configured API key. Every connection in the graph is tagged EXTRACTED (explicit in the source) or INFERRED (resolved by Graphify), so you can always tell what was read directly from what was inferred. Crucially, Graphify is not a vector index: there are no embeddings and no vector store, just a real graph you can traverse, trace paths through, and explain concepts from. For teams that want the graph always live across their SDLC, Graphify Cloud adds monorepo and cross-repo support, formal verification, code review with a bird’s-eye view, and connectors for Sentry and Jira.
EXTRACTED versus INFERRED tags make each relationship auditable./graphify command works in four major AI coding assistants.Freemium. The open-source CLI is free forever under Apache 2.0. Graphify Cloud, the hosted always-on memory service, starts with a free 14-day trial and then moves to paid plans.
Developers and teams onboarding onto large or legacy codebases, reviewers who want a bird’s-eye view of every change, and anyone tired of losing project context between agent sessions — a persistent graph remembers what chat windows forget.