google-agents-cli-scaffold

The `google-agents-cli-scaffold` skill is designed to streamline the creation and enhancement of agent projects within the Google ADK (Agent Development Kit) ecosystem. It addresses the challenges of setting up new agent projects, integrating CI/CD pipelines, managing deployment targets, and scaffolding infrastructure without requiring users to manually configure project directories or write repetitive boilerplate code. By providing a structured CLI-based workflow, this skill ensures consistency, reduces setup time, and supports both prototype-first and fully deployable project structures.

The main capabilities of this skill include scaffolding new projects via the `agents-cli scaffold create` command, enhancing existing projects with deployment and CI/CD integrations using `scaffold enhance`, and upgrading project configurations with `scaffold upgrade`. It supports multiple agent architectures, such as RAG (retrieval-augmented generation) with vector or document search, A2A protocols, and prototype projects. Users can specify deployment targets like agent runtime, Cloud Run, or GKE, configure CI/CD runners (GitHub Actions or Cloud Build), and select session storage options. Additionally, the skill provides guidance for mapping user terminology to the correct CLI flags and integrates seamlessly with the prototype-first workflow.

This skill is ideal for developers, AI engineers, and technical teams working with the ADK who need to quickly initialize projects or improve existing ones. Typical use cases include creating a new agent project from scratch, adding deployment capabilities, integrating CI/CD pipelines, or upgrading the infrastructure and configuration of existing projects. It is particularly useful for teams adopting a prototype-first approach, allowing them to validate ideas quickly before moving to full deployment, while ensuring compatibility with the Gemini Enterprise Agent Platform and related tools.