Live Demo: research-agent-nine-gamma.vercel.app
A premium, high-performance research assistant interface powered by Google Gemini, LangChain, and MCP. This application provides a seamless, multi-modal chat experience designed for deep analysis, document reasoning, and real-time research.
- 🧠 Multi-Modal Reasoning: Upload and analyze images, PDFs, text documents, CSVs, JSON, and Word files directly in the chat.
- ⚡ Powered by Gemini 2.5: Choose between Gemini 2.5 Pro (Brain), Flash (Fast), and Flash-Lite (Lightweight) for tailored responses.
- 🔍 Real-time Research: Leveraging advanced agentic workflows to browse and synthesize information from across the web.
- 🎨 Premium UI/UX: Dark-themed, responsive design with glassmorphism, fluid animations (Framer Motion), and intuitive navigation.
- 🔒 Session Integrity: Securely initialize your session with your own API key—keys are never stored on the server.
- Frontend: React 19, TypeScript, Vite
- Styling: Chakra UI v3, Framer Motion (Animations)
- Content: React Markdown with GFM support
- Backend API: Hosted LangChain Agent via Hugging Face Space
- Core Logic: Google Gemini API, MCP (Model Context Protocol)
- Deployment: Vercel
- Node.js: v18.0 or higher
- Google AI API Key: Obtain yours from Google AI Studio.
-
Clone the repository:
git clone https://github.com/Sauravroy34/ResearchAgent.git cd ResearchAgentUI -
Install dependencies:
npm install
-
Start the development server:
npm run dev
- Initialization: On the first launch, enter your Google API Key and select your preferred Gemini Model.
- Chat: Start asking questions! You can use the 📎 icon to attach files for analysis.
For more info about the mcp server and backend used check out my this project https://github.com/Sauravroy34/ResearchPaperMCP