TL;DR
A CLI and skill set that turns coding assistants into experts for building, evaluating, deploying, and managing AI agents on Google Cloud (Gemini Enterprise Agent Platform).
Key features
Agent Development Lifecycle Support: Supports the entire development process, including code preservation rules, model selection, and project scaffolding (create, enhance, upgrade).
Integrated Evaluation Methodology: Provides systematic agent evaluation using LLM-as-judge, adaptive rubrics, metrics, and datasets.
Cloud Deployment & Management: Supports deployment to Agent Runtime, Cloud Run, GKE, CI/CD, secrets management, Gemini Enterprise registration, and observability via Cloud Trace.
Universal Coding Agent Integration: Works seamlessly with various coding agents like Antigravity CLI, Claude Code, and Codex, extending their expertise.
When to use it
Useful for developers who need to build and deploy AI agents on Google Cloud.
Use when you want to enhance agent development, evaluation, and deployment expertise while using coding agents like Claude Code or Codex.
Suitable for standardizing the entire workflow from development to monitoring of agents in an enterprise environment.