What problem does it solve? Setting up a new Google ADK agent project involves many decisions—templates, deployment targets, CI/CD runners, session storage, and datastores—and doing it by hand is error-prone. This Skill guides the AI through the agents-cli scaffold commands so projects are created, enhanced, or upgraded correctly the first time. ## Core Features & Use Cases - Project Creation: Run agents-cli scaffold create with the right flags for agent templates (adk, agentic_rag), deployment targets (agent_runtime, cloud_run, gke), and prototype mode. - Enhancement & Upgrade: Add deployment or CI/CD to an existing project with scaffold enhance, or migrate to a newer CLI version with scaffold upgrade while preserving customizations. - Reference Scaffolding: Generate temporary reference projects in /tmp/ to cherry-pick Terraform, Dockerfile, or CI/CD workflow files for non-standard project structures. - Use Case: A developer says "build me a Python agent that exposes A2A and deploys to Cloud Run"—the Skill clarifies requirements, then runs agents-cli scaffold create my-a2a-agent --agent adk --deployment-target cloud_run --prototype to produce a valid project without hand-writing A2A code. ## Quick Start Ask the assistant to create a new ADK agent project with your chosen deployment target, and it will clarify your requirements before running the appropriate agents-cli scaffold command.
