Summary

GraphRAG Workbench is an interactive web application designed to transform documents into explorable 3D knowledge graphs using Microsoft's GraphRAG framework. It targets engineers, researchers, and power users seeking advanced AI-powered analysis and immersive visual analytics for their data. The project offers a benefit by enabling users to intuitively navigate, query, and understand complex relationships within document-derived knowledge graphs.

How It Works

The application leverages Microsoft's GraphRAG pipeline for robust entity extraction, relationship mapping, and community detection from uploaded documents, primarily PDFs. It then renders these extracted knowledge graphs in an interactive 3D environment using React Three Fiber and Three.js, facilitating immersive exploration. A key differentiator is the integrated AI chat interface, allowing users to query the knowledge graph using natural language, complemented by features like community isolation and smart node sizing for deeper insights.

Quick Start & Requirements

• Installation: Clone the repository, run npm install (or pnpm install), set up the .env file with an OPENAI_API_KEY, install the GraphRAG Python package (pip install graphrag), and start the development server with npm run dev.
• Prerequisites: Node.js 18+, Python 3.10+ (for GraphRAG backend), and an OpenAI API key.
• Access: Navigate to http://localhost:3000 in your browser.
• Docs: Usage Guide is detailed within the README.

Highlighted Details

• Interactive 3D Visualization: Features smooth navigation, community detection with color-coded boundaries, smart node sizing based on centrality, advanced filtering, and real-time search with highlighting.
• Document Management: Supports drag-and-drop PDF uploads, batch processing, archive management for saving/restoring graph versions, and real-time indexing progress tracking.
• AI-Powered Querying: Integrates a chat interface for natural language queries and provides AI-generated community reports.
• Advanced Features: Includes a Community Isolator for focused analysis, relationship weighting visualization, and bloom effects for enhanced visual appeal.

Maintenance & Community

The provided README does not detail specific contributors, community channels (like Discord or Slack), or a public roadmap. Contributions are welcomed via pull requests.

Licensing & Compatibility

This project is licensed under the MIT License, which is permissive and generally suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

The workbench currently supports only PDF file uploads. It requires an active OpenAI API key for its AI functionalities. Users may experience slow 3D performance with highly complex graphs or memory issues when processing very large documents, and WebGL 2.0 support is mandatory for the 3D visualization.