Graphify-Labs/graphify

Health v2 · maintenance only

Flags: young

Adoption not part of the score

111080 stars · 10801 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Graphify is an open-source (Apache 2.0) Python CLI and /graphify skill that turns a codebase—including docs, SQL schemas, configs, and PDFs—into a queryable knowledge graph using deterministic local tree-sitter AST parsing, with no vector store or embeddings. It integrates with AI coding assistants like Claude Code, Cursor, Codex, and Gemini CLI, and exposes the graph via a CLI and an MCP server, returning answers as explicit paths with file:line citations.

Use cases

• map my codebase into a knowledge graph for my AI assistant
• query who owns billing without grepping through files
• understand how auth connects to the database in a large repo
• generate an architecture report and interactive graph of a repository
• give Claude Code or Cursor grounded code navigation with citations
• index docs, SQL schemas, and Terraform alongside source code
• run codebase retrieval fully on-device with no telemetry

When to choose

• you want deterministic, auditable code retrieval with file:line citations instead of fuzzy vector search
• you use an AI coding assistant (Claude Code, Cursor, Codex, Gemini CLI, etc.) and want it to traverse real call/import edges
• you need on-device parsing with no code leaving your machine and no telemetry
• you want to index non-code assets like Markdown, PDFs, SQL schemas, and Terraform into one graph
• you need graph export to Neo4j, FalkorDB, Obsidian, or GraphML

When to avoid

• you need semantic embedding-based retrieval over unstructured natural-language corpora rather than code structure
• you want a hosted SaaS with team features out of the box (the enterprise layer is early access)
• your project is in a language outside the 36 supported tree-sitter grammars
• you need real-time incremental indexing of very rapidly changing code at large scale

Facets

cli-tool · maturity active

parser search-engine rag mcp developer-tools cli code-review developer-tools large-language-models apis python cli cross-platform windows self-hosted knowledge-graph graphrag tree-sitter ast-parsing code-analysis code-search claude-code cursor codex gemini-cli mcp-server on-device no-vector-store leiden-clustering ai-coding-assistants ai-agents retrieval-augmented-generation command-line macos linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Graphify-Labs/graphify")