KResearch is an AI-powered application designed for in-depth research, generating comprehensive reports on complex topics by synthesizing information from various web sources. It targets students, analysts, and professionals seeking to quickly gain a deep understanding of subjects, providing well-structured reports with citations and knowledge graphs.

How It Works

The application employs a multi-agent AI system, featuring 'Alpha' (Strategist) and 'Beta' (Tactician) agents that collaborate to devise and execute research plans. It operates through iterative cycles of planning, searching (via Google Search), and reading, ensuring comprehensive data gathering. The process is visualized in real-time, and outputs include Markdown reports with source citations and Mermaid.js knowledge graphs.

Quick Start & Requirements

• Installation: Clone the repository (git clone https://github.com/KuekHaoYang/KResearch.git), navigate into the directory (cd KResearch), and install dependencies (npm install).
• Prerequisites: A Google Gemini API key is mandatory. Node.js (LTS recommended) and a package manager (npm or yarn) are also required.
• Usage: Run the development server with npm run dev and access the application via the provided local address (e.g., http://localhost:5173).
• Docker: Pull and run the pre-built image: docker run -p 8080:80 --name kresearch kuekhaoyang/kresearch:latest. Access at http://localhost:8080.
• Configuration: API key can be entered in the in-app Settings modal (recommended for Docker) or via a .env file for local development.
• Documentation: Project GitHub Repository

Highlighted Details

• Utilizes a multi-agent AI system with distinct Strategist and Tactician roles.
• Offers configurable research modes: 'Balanced', 'Deep Dive', 'Fast', and 'Ultra Fast'.
• Generates both Markdown reports with citations and visual knowledge graphs using Mermaid.js.
• Features a modern, responsive UI built with React and Tailwind CSS, supporting light and dark modes.

Maintenance & Community

The project is actively maintained by Kuek Hao Yang. For contributions, questions, or feature requests, users are directed to the GitHub Issues page.

Licensing & Compatibility

Distributed under the MIT License, allowing for broad use and modification, including commercial applications.

Limitations & Caveats

The application's functionality is entirely dependent on the Google Gemini API, requiring a valid API key. The effectiveness and accuracy of the generated reports are subject to the capabilities and limitations of the underlying AI model.