copilotkit__copilotkit/examples/integrations/ms-agent-framework-python/README.md

CopilotKit <> Microsoft Agent Framework (Python)

This is a starter template for building CopilotKit experiences using the Microsoft Agent Framework. It ships with a Next.js UI and a FastAPI server that exposes a Microsoft Agent Framework agent over the AG-UI protocol, so you can study and customize both sides of the stack.

Prerequisites

• OpenAI or Azure OpenAI credentials (for the Microsoft Agent Framework agent)
• Python 3.12+
• uv
• Node.js 20+
• Any of the following package managers:
  • npm (default)
  • pnpm
  • yarn
  • bun

Getting Started

1. Install dependencies using your preferred package manager:

   # Using npm (default)
   npm install
   
   # Using pnpm
   pnpm install
   
   # Using yarn
   yarn install
   
   # Using bun
   bun install
   

   Note: This automatically sets up the Python environment as well.

   If you have manual issues, you can run:

   npm run install:agent
   

2. Set up your agent credentials. The backend automatically uses Azure when the Azure env vars below are present; otherwise it falls back to OpenAI. Create a .env file inside the agent folder with one of the following configurations:

   OpenAI

   OPENAI_API_KEY=sk-...your-openai-key-here...
   OPENAI_CHAT_MODEL_ID=gpt-4o-mini
   

   Azure OpenAI

   AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
   AZURE_OPENAI_CHAT_DEPLOYMENT_NAME=gpt-4o-mini
   # If you are not relying on az login:
   # AZURE_OPENAI_API_KEY=...
   

3. Start the development server:

   # Using npm (default)
   npm run dev
   
   # Using pnpm
   pnpm dev
   
   # Using yarn
   yarn dev
   
   # Using bun
   bun run dev
   

   This will start both the UI and the Microsoft Agent Framework server concurrently.

CopilotKit Intelligence & Threads (Optional)

CopilotKit Intelligence provides durable, multi-turn conversation threads with memory persistence. To enable:

Prerequisites

• Docker Desktop or equivalent
• CopilotKit License Token (get from CopilotKit Cloud)
• Intelligence Repository - Clone the CopilotKit Intelligence repository locally

Setup

1. Clone the Intelligence repository (if not already cloned):

# Clone to a sibling directory (recommended)
cd ..
git clone https://github.com/CopilotKit/Intelligence.git
cd -

2. Start the intelligence stack:

# Set your license token (if not already in .env)
echo "COPILOTKIT_LICENSE_TOKEN=your-token-here" >> .env

# Set the path to your Intelligence repo checkout
# Default: ../../../Intelligence (assumes CopilotKit monorepo layout)
# For standalone starters, adjust to your actual path:
export INTELLIGENCE_REPO=../Intelligence  # if cloned as sibling

# Start intelligence services (postgres, redis, intelligence API)
docker compose -f docker-compose.intelligence.yml up -d

# Verify services are healthy
docker compose -f docker-compose.intelligence.yml ps

3. Ensure intelligence configuration in .env (uncomment if already present):

COPILOTKIT_LICENSE_TOKEN=your-token-here
INTELLIGENCE_API_URL=http://localhost:4201
INTELLIGENCE_GATEWAY_WS_URL=ws://localhost:4401

4. Run your application:

npm run dev

The intelligence stack will now handle conversation threads, state persistence, and memory. See docker-compose.intelligence.yml and docker/01-create-databases.sql for configuration details.

Note: The Intelligence service is built from source and requires the Intelligence repository. If you're using this as a standalone starter (extracted from the CopilotKit monorepo), you must clone the Intelligence repo separately and set INTELLIGENCE_REPO to point to it.

Stopping Intelligence

# Stop services (keeps data)
docker compose -f docker-compose.intelligence.yml stop

# Stop and remove containers + volumes (fresh start)
docker compose -f docker-compose.intelligence.yml down -v

Available Scripts

The following scripts can also be run using your preferred package manager:

• dev – Starts both UI and agent servers in development mode
• dev:debug – Starts development servers with debug logging enabled
• dev:ui – Starts only the Next.js UI server
• dev:agent – Starts only the Microsoft Agent Framework server
• build – Builds the Next.js application for production
• start – Starts the production server
• lint – Runs ESLint for code linting
• install:agent – Installs Python dependencies for the agent

Documentation

The main UI component is in src/app/page.tsx. You can:

• Modify the theme colors and styling
• Add new frontend actions
• Customize the CopilotKit sidebar appearance

📚 Documentation

• Microsoft Agent Framework – Learn more about Microsoft Agent Framework and its features
• CopilotKit Documentation – Explore CopilotKit’s capabilities
• Next.js Documentation – Learn about Next.js features and API

Contributing

Feel free to submit issues and enhancement requests! This starter is designed to be easily extensible.

License

This project is licensed under the MIT License – see the LICENSE file for details.

Troubleshooting

Agent Connection Issues

If you see "I'm having trouble connecting to my tools", make sure:

1. The Microsoft Agent Framework agent is running on port 8000
2. Your OpenAI/Azure credentials are set correctly
3. Both servers started successfully

Python Dependencies

If you encounter Python import errors:

cd agent
uv sync
uv run src/main.py