╔══════════════════════════════════════╗
║ ⚡ ⚡ ║
║ 🤖 AI Agent Orchestration 🎭 ║
║ ⚡ ⚡ ║
║ ┌─────────────┐ ┌─────────────┐ ║
║ │ Image │ │ Story │ ║
║ │ Generator │ │ Teller │ ║
║ │ 🎨 │ │ 📖 │ ║
║ └─────────────┘ └─────────────┘ ║
║ │ │ ║
║ └──────┬─────────┘ ║
║ │ ║
║ ┌─────────────┐ ║
║ │ Orchestrator│ ║
║ │ 🎯 │ ║
║ └─────────────┘ ║
║ ⚡ ⚡ ║
╚══════════════════════════════════════╝
Before starting, you need to configure a couple of variables in config.py
:
- Google Cloud Storage bucket for images 🗄️ - Update
GCS_IMAGE_BUCKET
with your workshop bucket - Toolbox endpoint 🔧 - Update
TOOLBOX_ENDPOINT
with the Cloud Run URL from thepar-devfest-sfeir
Google Cloud Project
First, create and activate a virtual environment:
python -m venv .venv
source .venv/bin/activate
To activate it in future sessions:
source .venv/bin/activate
Ensure you have the required dependencies:
pip install -r requirements.txt
From the solution (or workshop) folder:
uv run a2a_image_generator_agent/
From the solution (or workshop) folder:
uv run a2a_storyteller_agent/
From the solution (or workshop) folder:
adk web
To test and debug individual agents, you can use the a2a-inspector tool, which provides a user-friendly interface for interacting with your agents.
# Clone the a2a-inspector repository
git clone https://github.com/a2aproject/a2a-inspector.git
cd a2a-inspector
# Follow the installation instructions in the repository
- Start your agent server (e.g.,
uv run a2a_image_generator_agent/
) - Launch a2a-inspector and connect to your agent's endpoint
- Send test messages to verify agent behavior and responses
- Monitor logs to troubleshoot any issues