Agents CLI 🤖 Agent Free

What it does

Agents CLI is Google's official command-line tool and skill suite that turns a general-purpose coding assistant (such as Claude Code or Google's own Antigravity CLI) into a specialist for creating, evaluating and deploying AI agents on Google Cloud. Rather than being a coding agent itself, it's an add-on that teaches existing coding assistants a specific, well-scoped workflow: building on Google's Agent Development Kit (ADK).

What is inside

• Agent scaffolding and project creation for new ADK-based agents
• Evaluation tooling with metrics and grading to test agent behavior before shipping
• Deployment helpers targeting Cloud Run, GKE and Agent Runtime
• Integration with the Gemini Enterprise Agent Platform
• Observability/monitoring tooling and CI/CD pipeline setup

Works with

Designed to be invoked by coding assistants generally (Claude Code and Google's Antigravity CLI are both mentioned), rather than being tied to one specific assistant.

How to install or connect

uvx google-agents-cli setup

Requires Python 3.11+, the uv package manager, and Node.js. It is also published on PyPI as google-agents-cli.

Maintenance and safety

Apache-2.0 licensed and officially maintained by Google (github.com/google/agents-cli). With roughly 6,000 stars and 670 forks, it is a young but credible project backed by a major cloud vendor, and it deploys to real, billable Google Cloud resources (Cloud Run, GKE), so review what it provisions before running it against a production project or billing account.

Who should use it

Teams already building on Google Cloud who want their existing coding assistant to also handle ADK agent scaffolding, evaluation and deployment correctly, instead of hand-writing Cloud Run/GKE deployment configuration and evaluation harnesses from scratch. Less useful for developers not targeting Google Cloud specifically, since its value is tied to that ecosystem. It's best understood as a productivity multiplier layered on top of whatever coding assistant you already use, rather than a replacement for one, and is a natural fit for teams that already standardize on Google Cloud for other infrastructure and want consistent ADK conventions across projects.