This workshop showcases the new Amazon Bedrock Agents feature multi-agent collaboration capabilities through an Energy Efficiency Management System. The system consists of a supervisor agent that orchestrates two specialized sub-agents, each handling specific aspects of energy management and customer service.
├── 1- Forecasting Agent
├── 2- Solar Panel Agent
├── 3- Energy Efficiency Management Agent (Supervisor)
└── 4- Clean up
The supervisor agent coordinates the activities of three specialized sub-agents, routing customer queries and requests to the appropriate agent while maintaining context and ensuring seamless interactions. The architecture looks as following:
- Provides current energy consumption data
- Provides consumption forecasts
- Provides user consumption statistics
- Contains code interpretation capabilities to analyze forecasting data
- Provides installation guidelines and requirements
- Offers maintenance instructions and schedules
- Enables support ticket creation
- Tracks existing support tickets
- Forecast agent setup
- Solar panel agent setup
- Multi-agent collaboration setup
- Supervisor agent invocation
- Clear up
- AWS Account with appropriate permissions
- Amazon Bedrock access
- Basic understanding of AWS services
- Python 3.8+
- Latest Boto3 SDK
- AWS CLI configured
- Clone this repository
git clone https://github.com/aws-samples/bedrock-multi-agents-collaboration-workshop.git- Install dependencies (Not required if you are using Jupyterlab in Sagemaker Studio)
pip install -r requirements.txt- Follow the setup instructions in the workshop guide
See CONTRIBUTING for more information.
This project is licensed under the MIT-0 License - see the LICENSE file for details.
For support and questions, please open an issue in the repository.
Note: This workshop is for educational purposes and demonstrates the capabilities of Amazon Bedrock Agents' multi-agent collaboration feature.