This project demonstrates a comprehensive research assistant implementation using AG-UI protocol, LangGraph, and CopilotKit. The system allows users to ask research questions and receive detailed reports with sources.

The project consists of two main components:

Python-based backend that:

• Processes research queries
• Performs web searches using Serper API
• Generates detailed research reports using OpenAI
• Implements AG-UI protocol with FastAPI
• Uses LangGraph for the research workflow

Next.js/React frontend that:

• Provides a user-friendly research interface
• Connects to the backend using CopilotKit
• Visualizes research progress with real-time updates
• Displays formatted research reports with sources
• Implements responsive UI for research workflow

This project demonstrates AG-UI protocol implementation:

• Event-driven architecture for real-time updates
• State management using snapshots and deltas
• Standardized message events for communication
• Progress tracking through state updates

1. Navigate to the backend directory:

   cd ag-ui-research-agent

2. Install dependencies:

   poetry install

3. Create a .env file with API keys:

   OPENAI_API_KEY=your-openai-key
   SERPER_API_KEY=your-serper-key
   

4. Run the backend:

   poetry run python -m src.my_endpoint.main

1. Navigate to the frontend directory:

   cd ag-ui-research-frontend

2. Install dependencies:

   npm install

3. Start the development server:

   npm run dev

4. Open http://localhost:3000 to access the application

┌─────────────────┐       AG-UI       ┌─────────────────┐
│                 │      Protocol      │                 │
│  React Frontend │<----------------->│ Python Backend  │
│   (CopilotKit)  │     (HTTP/SSE)    │   (LangGraph)   │
└─────────────────┘                    └─────────────────┘
        │                                       │
        │                                       │
┌─────────────────┐                     ┌─────────────────┐
│                 │                     │                 │
│    User         │                     │   Web Search    │
│   Interface     │                     │   (Serper API)  │
└─────────────────┘                     └─────────────────┘
                                                │
                                        ┌─────────────────┐
                                        │                 │
                                        │  Report         │
                                        │  Generation     │
                                        │  (OpenAI)       │
                                        └─────────────────┘

• Natural language research query processing
• Real-time research progress tracking
• Structured research reports with sections
• Source management and citation
• Modern, responsive UI
• Streaming updates using AG-UI protocol

• Backend: Python, FastAPI, LangGraph, OpenAI, Serper API
• Frontend: Next.js, React, CopilotKit
• Protocol: AG-UI (Agent User Interaction Protocol)

[Specify license or state that it's private/internal]