GraphWeaver

GraphWeaver is a high-performance Graph RAG (Retrieval-Augmented Generation) knowledge engine. It leverages the power of "Entity-Relationship" modeling and "Semantic Vector" search to solve complex multi-hop reasoning challenges that traditional RAG systems often struggle with.

Click here to try the Demo

---

Key Features

• Hybrid Retrieval Engine: Combines Vector Search (for semantic entry points) with Graph Diffusion (for logical relationship reasoning).
• Containerized Deployment: Easy setup with Docker Compose for local development and testing.
• High Performance: Backend implemented in Go (Golang) for efficient concurrency and low-latency processing.
• Modern Knowledge Management: Automated entity and relationship extraction from unstructured data (PDF/Markdown).
• Interactive UI: Sleek and responsive dashboard built with React, TypeScript, and Tailwind CSS.

---

Technology Stack

💻 Frontend & Styling

⚙️ Backend & Core

🗄️ Databases & Storage

🚀 Infrastructure & DevOps

---

System Architecture

GraphWeaver follows a monolithic architecture with clear separation of concerns, designed for ease of development and deployment via Docker Compose.

---

Quick Start

Prerequisites

• Docker & Docker Compose
• Go 1.24+
• Make (optional, for convenience)

Running with Docker Compose

1. Clone the repository:

   git clone https://github.com/suyw-0123/GraphWeaver.git
   cd GraphWeaver

2. Configure your environment:

   cp .env.example .env
   # Edit .env and add your GEMINI_API_KEY

3. Start the services:

   docker-compose up -d

4. Access the web interface at http://localhost:80.

---

Development

We use a Makefile to automate common development tasks:

• make dev-tools: Install necessary Go development tools.
• make deps: Download dependencies.
• make fmt: Format code.
• make lint: Run linters.
• make test: Run unit and integration tests.
• make docker-build: Build Docker images locally.

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.