Django CopilotKit Assistant

A Django application that integrates CopilotKit with LangGraph agents for AI-powered assistance.

Features

• 🤖 AI-powered chat agent using LangGraph
• 🔍 Web search capabilities with Tavily
• ⏰ Current datetime tool
• 🌐 RESTful API endpoints
• 🔄 Streaming responses

Prerequisites

• Python 3.12
• Virtual environment (recommended)
• API keys for the services you want to use

Setup

1. Clone and navigate to the project:

   cd assistant

2. Create your virtual environment:

   # Windows
   py -m venv env
   # Windows (if you have more than one python version installed, create virtual environment using specific verion)
   py -3.12 -m venv env
   
   # macOS/Linux
   source env/bin/activate

3. Activate your virtual environment:

   # Windows
   env\Scripts\activate
   
   # macOS/Linux
   source env/bin/activate

4. Install dependencies:

   pip install -r requirements.txt

5. Set up environment variables:

   Create a .env file:

   # Create .env file with your API keys
   DEEPSEEK_API_KEY=your_deepseek_api_key_here
   OPENAI_API_KEY=your_openai_api_key_here
   TAVILY_API_KEY=your_tavily_api_key_here
   DEBUG=True
   SECRET_KEY=your_django_secret_key

6. Run database migrations (Optional):

   python manage.py migrate

7. Start the server:

   uvicorn assistant.asgi:application --reload

Example Nextjs: app/api/copilotkit/route.ts

import { NextRequest } from "next/server";
import {
  CopilotRuntime,
  EmptyAdapter,
  copilotRuntimeNextJSAppRouterEndpoint,
} from "@copilotkit/runtime";

const serviceAdapter = new EmptyAdapter();

const runtime = new CopilotRuntime({
  remoteEndpoints: [
    {
      url: `${process.env.NEXT_PUBLIC_DJANGO_SERVER_URL || "http://localhost:8000"}/api/copilotkit`,
    },
  ],
});

export async function POST(req: NextRequest) {
  const { handleRequest } = copilotRuntimeNextJSAppRouterEndpoint({
    runtime,
    serviceAdapter,
    endpoint: "/api/copilotkit",
  });

  return handleRequest(req);
}

API Endpoints

CopilotKit Endpoints

• GET/POST /api/copilotkit/info - Get agent information
• POST /api/copilotkit/agent/{name} - Execute an agent
• POST /api/copilotkit/agent/{name}/state - Get agent state
• POST /api/copilotkit/action/{name} - Execute an action

Utility Endpoints

• GET /api/health - Health check
• POST /api/echo - Echo message (for testing)

Usage Examples

Test the info endpoint:

curl -X POST http://localhost:8000/api/copilotkit/info

Test the health endpoint:

curl http://localhost:8000/api/health

Troubleshooting

500 Internal Server Error

If you're getting 500 errors, check:

1. API Keys: Make sure your API keys are set in the .env file
2. Environment Variables: Verify the .env file is being loaded
3. Dependencies: Ensure all packages are installed correctly

Common Issues

• Missing API Keys: The application will show helpful error messages if API keys are missing
• Network Issues: Make sure you have internet access for API calls
• Port Conflicts: If port 8000 is busy, use a different port: uvicorn assistant.asgi:application --host 0.0.0.0 --port 8001 --reload

Development

Project Structure

assistant/
├── app/
│   ├── agent.py          # LangGraph agent definition
│   ├── chat.py           # Chat node implementation
│   ├── copilotkit_integration.py  # Django-CopilotKit integration
│   ├── model.py          # LLM configuration
│   ├── schema.py         # Agent state schema
│   ├── sdk.py            # CopilotKit SDK setup (define your agent here)
│   └── views.py          # Django views
├── assistant/
│   ├── settings.py       # Django settings
│   └── urls.py           # Main URL configuration
├── manage.py             # Django management script
├── requirements.txt      # Python dependencies
└── setup_env.py          # Environment setup script

Adding New Tools

To add new tools to the agent:

1. Define the tool in app/chat.py
2. Add it to the tools list in the get_tools function
3. Update the system message if needed

Adding New Agents

To add new agents:

1. Create a new graph in app/agent.py
2. Add the agent to the SDK configuration in app/sdk.py
3. Update the URL patterns if needed