A modular, production-grade AI Agent orchestration framework engineered using the Google ADK (Agent Development Kit), FastAPI, Pydantic v2, and asynchronous event-driven execution.
This microservice provides a resilient agent orchestration interface featuring clean architectural decoupling, structured schema validation, and Multi-Agent patterns.
- Multi-Agent & Tool Calling: Implements advanced agentic loops capable of delegating tasks and autonomously invoking tools using the latest ADK Runner pattern.
- RESTful Orchestration: Powered by FastAPI with an interactive Swagger UI for real-time testing, acting as a scalable backend for LLM interactions.
- Hexagonal Design (Ports & Adapters): Strict separation between Domain, Application, and Infrastructure layers, ensuring horizontal scalability.
- Non-blocking Execution: Full
asyncioimplementation to handle high-throughput I/O and event streams efficiently. - Data Integrity: Pydantic v2 ensures strict typing and data compliance across all agent-tool LLM payloads.
The service is containerized for seamless reproduction.
# Clone the repository
git clone https://github.com/gabaoun/CabaModel.git
cd CabaModel
# Boot the service via Docker Compose
docker-compose up -d --buildThe interactive API documentation will be available at http://localhost:8000/docs.