Assignment 2 starter kit for UCLA Extension's Agentic AI course. Students build a Google Workspace Assistant using Google's Agent Development Kit (ADK).
Sign in to like and favorite skills
Assignment 2 starter kit for UCLA Extension's Agentic AI course. Students build a Google Workspace Assistant using Google's Agent Development Kit (ADK).
workspace_assistant/ ├── agent.py # Main agent definition (students complete) ├── main.py # Entry point ├── tools/ │ ├── auth.py # Google OAuth helper │ ├── calendar_tools.py # Option A stubs │ ├── tasks_tools.py # Option B stubs │ └── mcp_tools.py # Part 2: GitHub MCP integration ├── config/ │ ├── settings.py # Configuration │ ├── mcp_servers.json # MCP server config (optional) │ └── credentials/ # OAuth files (gitignored) ├── tests/ │ └── test_mcp.py # MCP integration test └── docs/ ├── google_auth_setup.md ├── github_mcp_setup.md └── reflection_template.md
Part 1 (70 pts): Choose ONE option and implement:
Part 2 (30 pts): GitHub MCP integration:
Bonus (+25 pts): Tool search with defer_loading
# Setup conda create -n workspace-agent python=3.11 -y conda activate workspace-agent pip install -r requirements.txt cp .env.example .env # Run python main.py --interactive # Test with ADK Web UI adk web