Overview

Graphify is a revolutionary AI coding assistant skill that transforms any codebase, documentation, or data into a queryable knowledge graph. By ingesting everything from code files, SQL schemas, R scripts, shell scripts, to even documents, papers, images, and videos, Graphify creates a rich, interconnected understanding of your project that you can query directly. Say goodbye to endless grep sessions and hello to interactive exploration of your entire development landscape.

Key Features

• Unified Knowledge Graph: Seamlessly integrates application code, database schemas, and infrastructure definitions into a single, queryable graph.
• Local-First Code Parsing: Code is analyzed using deterministic tree-sitter AST parsing, ensuring privacy and speed with no LLM involvement and no data leaving your machine.
• Semantic Pass for Other Data: While code is processed locally, documents, PDFs, images, and videos can leverage your AI assistant's model or a configured API key for semantic understanding, all mapped into the same graph.
• Transparent Edge Explanations: Every connection within the graph is tagged as EXTRACTED (explicitly present in the source) or INFERRED (resolved by Graphify), providing clear insights into how relationships were discovered.
• Not a Vector Index: Operates on a true graph structure, enabling precise traversal, pathfinding, and explanation of concepts, rather than relying on approximate vector embeddings.
• Interactive Visualization: Generates graph.html, an interactive graph visualization that allows users to click through nodes, filter, and search for concepts within their project.
• Actionable Reports: Produces GRAPH_REPORT.md with key concepts, surprising connections, and suggested queries, alongside graph.json for programmatic querying.
• Broad AI Assistant Compatibility: Works seamlessly with popular AI coding assistants including Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, and over 15 others.

Typical Use Cases

• Accelerated Onboarding: New team members can quickly grasp the architecture and interdependencies of a complex project by querying the knowledge graph.
• Deep Code Understanding: Developers can explore "god nodes" (most connected concepts), identify subsystems through community detection, and understand how different parts of the codebase interact.
• Debugging and Troubleshooting: Trace the path between related components to pinpoint the root cause of issues across code, data, and infrastructure.
• Architecture Review: Gain a comprehensive overview of your project's structure, identifying potential bottlenecks or surprising links between seemingly unrelated modules.
• Documentation and Knowledge Management: Centralize and query all project knowledge, including code, documentation, and research papers, in one unified interface.
• Cross-Language Analysis: Understand relationships across nearly 40 programming languages thanks to robust AST parsing.
• Investigating Legacy Code: Unravel the logic and dependencies of older codebases without needing to manually sift through vast amounts of files.
• AI Assistant Enhancement: Empower your AI assistant to provide more contextually relevant and accurate answers by giving it direct access to a structured understanding of your project.