research-weaver

# Research Weaver 🔬

An AI-powered multi-agent research system that conducts comprehensive, iterative research by intelligently searching, reading, and synthesizing information from multiple sources.

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
[![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
[![Streamlit](https://img.shields.io/badge/Streamlit-1.29+-red.svg)](https://streamlit.io/)

[简体中文](./README_CN.md) | English

## ✨ Features

### 🤖 Multi-Agent Architecture
- **Supervisor Agent**: Orchestrates research by creating specialized researcher agents
- **Researcher Agents**: Focused agents that handle specific research aspects
- **Dynamic Agent Creation**: Automatically spawns new researchers based on discovered topics

### 🔍 Intelligent Research Process
- **Iterative Search**: Continuously refines search queries based on findings
- **Multi-Source Integration**: Gathers information from various web sources
- **Knowledge Gap Detection**: Identifies missing information and creates targeted sub-researches
- **Citation Management**: Tracks and formats all sources with proper citations

### 💡 Advanced Capabilities
- **LLM-Powered Analysis**: Uses AI to understand context and synthesize information
- **Memory Management**: Persistent storage for research sessions with caching
- **Real-time Progress Tracking**: Visual interface shows live research progress
- **Structured Reports**: Generates comprehensive reports with clear sections and citations

### 🎯 Key Differentiators
- **Truly Autonomous**: Agents make independent decisions about what to research next
- **Context-Aware**: Each agent understands the overall research goal and its specific role
- **Scalable**: Can handle complex topics by breaking them into manageable sub-researches
- **Transparent**: Full visibility into the research process and decision-making

## 🚀 Quick Start

### Prerequisites