LeapRAG

<p align="center">
  <img src="https://github.com/user-attachments/assets/a9a7271b-677a-4463-afd0-9d3c58053453" alt="LeapRAGAgent Logo" />
</p>

<h1 align="center">LeapRAG</h1>

<p align="center">
  Dual-powered by RAG & A2A Protocol - Defining the Next Generation of Knowledge Agents
</p>

<p align="center">
  <a href="#quick-start">Self-Hosting</a> ·
  <a href="#key-features">Key Features</a>
</p>

<p align="center">
  <a href="./README.md"><img alt="English" src="https://img.shields.io/badge/English-d9d9d9"></a>
  <a href="./README_CN.md"><img alt="简体中文" src="https://img.shields.io/badge/简体中文-d9d9d9"></a>
</p>

LeapRAG is an open-source platform that combines state-of-the-art Retrieval Augmented Generation (RAG) capabilities with Google's A2A (Agent-to-Agent) protocol. The platform enables users to create knowledge-rich, intelligent agents that deliver highly accurate, context-aware responses backed by your own data sources. These agents are automatically compliant with the A2A protocol, making them discoverable and usable by any A2A-compatible client without additional development effort.

## Demo Video: From Model Setup to A2A Client Testing

This video demonstrates the complete workflow of LeapRAG, including:

- Large language model configuration
- Knowledge base creation
- Knowledge base chat testing
- A2A agent creation
- Testing with Google A2A UI client

**Note:** Sensitive information in the video has been redacted.

https://github.com/user-attachments/assets/69755681-e8a7-4283-ae70-834b209e2721

## Key Features

**1. User-Friendly RAG Platform**:

- Simple and intuitive user interface for quickly building high-quality RAG applications without technical background
- End-to-end visual workflow from document upload to knowledge base creation
- Pre-configured best practices that require no parameter adjustments in most scenarios
- What-you-see-is-what-you-get knowledge base testing with real-time validation
- Built-in tutorials and prompts to assist with each