Auto Research AI is a full-stack application that allows users to input any research query and receive high-quality, summarized answers using web search data and generative AI. The platform streamlines the research process by combining SERP results with LLM-based summarization.
- Framework: Express.js with TypeScript
- Database: PostgreSQL (hosted on Neon)
- ORM: Prisma
- APIs used:
- SerpAPI for real-time search results
- Gemini API (Google Generative AI) for summarization
- Framework: React.js
- Styling: Tailwind CSS
- Authentication: Clerk for secure user login and session management
- User authentication using Clerk
- Search anything and fetch relevant data via SerpAPI
- Summarize search results using Google s Gemini model
- Store all queries and summaries in a PostgreSQL database Auto Research AI - README
- View past queries with summaries in your dashboard
- Fast and modern UI built with React and Tailwind CSS
git clone https://github.com/Dipit12/Auto-Research-AI.git
cd auto-research-aicd server
npm install
npx prisma generate
npx prisma migrate dev --name init
node dist/indexCreate a .env file in the server/ directory with:
Auto Research AI - README
DATABASE_URL=your_neon_database_url
SERP_API_KEY=your_serp_api_key
GEMINI_API_KEY=your_gemini_api_key
cd client
npm install
npm run devThis project uses Clerk.dev for authentication.
Set the following variables in the frontend .env.local file:
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=your_key
CLERK_SECRET_KEY=your_secret_key
Search queries are processed through SerpAPI, and the results are summarized using the Gemini Pro API for optimal clarity and conciseness. Auto Research AI - README
- Frontend: Deployable to Vercel
- Backend: Deployable to Render, Railway, or Cloudflare Workers (if converted to Hono)
This project is licensed under the MIT License. Feel free to fork and contribute!