This project provides an AI-powered research assistant with a modern web interface, serving as an open-source alternative to proprietary services like OpenAI and Gemini. It's designed for users who need to conduct in-depth research on various topics, offering an interactive and visually appealing experience for controlling and monitoring the research process.

How It Works

The system combines search engines (via FireCrawl), web scraping, and large language models (via OpenAI) to perform iterative, deep research. It intelligently generates targeted search queries, creates follow-up questions to refine research needs, and processes multiple searches in parallel. Users can configure the depth and breadth of the research, with outputs delivered as detailed markdown reports including sources and real-time progress tracking.

Quick Start & Requirements

• Primary install / run command: npm install followed by npm run dev.
• Prerequisites: Node.js v14+, OpenAI API key, FireCrawl API key.
• Setup: Clone the repository, install dependencies, and configure API keys in .env.local. The application runs on http://localhost:3000.

Highlighted Details

• Modern web UI built with Next.js and shadcn/ui.
• Intelligent research process with recursive topic exploration.
• Configurable research depth and breadth parameters.
• Real-time progress tracking and downloadable markdown reports.

Maintenance & Community

The project is sponsored by AnotherWrapper. The original CLI project is by dzhng/deep-research.

Licensing & Compatibility

MIT License. Permissive for commercial use and integration into closed-source projects.

Limitations & Caveats

The project relies on external API keys for OpenAI and FireCrawl, which may incur costs. The NEXT_PUBLIC_ENABLE_API_KEYS=false setting is recommended for production deployments to manage API keys securely.