← All repos

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.

GitHubwww.graphify.com
ai-agentsantigravityastclaude-codecode-analysiscode-searchcodexcursordeveloper-toolsgeminigraphragknowledge-graphleidenllmmcpopenclawragskillstree-sitter
Browse cluster: Claude AI Agent Frameworks & MCP Tools →
1,393commits
208contributors
2languages

Tech stack & purpose

Graphify is a developer tool that converts codebases—including their documentation, SQL schemas, configurations, and PDFs—into queryable knowledge graphs. The project parses code deterministically using tree-sitter abstract syntax trees (ASTs) with no vector store, vector embeddings, or reliance on language models for code analysis. Created by Graphify Labs, a Y Combinator S26 company, graphify is available as a skill for multiple AI coding assistants including Claude Code, Cursor, Codex, and Gemini CLI. The tool's output consists of an interactive HTML graph visualization, a markdown report, and a queryable JSON graph file, each edge tagged to distinguish explicitly extracted connections from inferred relationships. Beyond free local code mapping, graphify optionally processes documentation, PDFs, images, and video using an LLM backend, and offers graph databases like Neo4j and FalkorDB as export targets, with installation available via PyPI as the `graphifyy` package.

Community & reference links

Languages

Python
100.0%
Dockerfile
0.0%

Contributors (top 30 of 208)