Graphify-Labs/graphify
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
137.1K
Loading…
···
Loading…
13.1K
1.6K
On GitHub
134
about 9 hours ago
about 9 hours ago
Overview
Graphify transforms codebases into queryable knowledge graphs using deterministic AST parsing. It creates a persistent graph of code, docs, and PDFs that AI agents can query instead of searching files. The tool emphasizes local processing, explainable edges, and integration with 20+ coding assistants.
Key Highlights
- Local-first parsing: Uses tree-sitter AST for deterministic code analysis without LLMs.
- Explainable connections: Every edge is tagged as EXTRACTED or INFERRED.
- Universal agent integration: Works with Claude, Cursor, Codex, and 20+ other assistants via a single install.
- Graph-based querying: Ask questions, trace paths, or explain concepts using the graph.json file instead of vector search.
Best For
Developers and teams building AI coding agents who need a reliable, local-first way to understand complex codebases and ensure their agents have accurate, traceable context.
Summary written from the project's README and homepage on Oct 19, 2026. Spotted something wrong? .
Releases
0 releases totalView all