google-agents-cli

What it is and what it does

google-agents-cli is a command-line toolchain designed to integrate with coding agents (Claude Code, Antigravity CLI, Codex, or similar) to streamline the full lifecycle of building and deploying agents on Google Cloud's Gemini Enterprise Agent Platform. It provides skills and CLI commands for scaffolding new agent projects, running local inference, evaluating agent behavior against datasets, grading traces, and deploying to Cloud Run, GKE, or Agent Runtime. The package bundles 15 runtime dependencies including google-cloud-aiplatform, cookiecutter for project templates, and rich for CLI output.

The tool is intended to reduce friction for coding agents learning how to build production agents—instead of requiring knowledge of multiple Google Cloud CLIs and services, a coding agent can ask agents-cli to handle scaffolding, evaluation, deployment, and observability. It also works as a standalone CLI for developers who prefer to run commands directly. Local development (create, run, eval) can use an AI Studio API key; cloud deployment requires Google Cloud credentials.

Use it for:

• Scaffold a new agent project with deployment and CI/CD templates using `agents-cli scaffold` or `agents-cli scaffold enhance`.
• Run agent inference over evaluation datasets and grade traces to measure quality before deployment.
• Deploy a trained agent to Google Cloud (Cloud Run, GKE, Agent Runtime) with a single command.
• Register and publish an agent with Gemini Enterprise for organizational use.
• Integrate agent-building workflows into a coding agent's skill set so it can autonomously scaffold and deploy.
• Set up observability (Cloud Trace, logging, third-party integrations) for production agents.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

A CLI toolchain that equips coding agents (Claude Code, Antigravity CLI, Codex) with commands and skills to scaffold, evaluate, deploy, and publish agents on Google Cloud's Gemini Enterprise Agent Platform.

Yes. The package is actively maintained, has no known vulnerabilities, and low install friction. Install it if you are building agents on Google Cloud or want a coding agent to handle agent scaffolding and deployment. Skip it if you are not targeting Gemini Enterprise Agent Platform or prefer manual Google Cloud CLI workflows.

Install

google-agents-cli on PyPI

pip install google-agents-cli

uv add google-agents-cli

poetry add google-agents-cli

Installing google-agents-cli

Before you install

Low friction: pure Python wheel with 15 runtime dependencies including well-maintained packages (click, PyYAML, google-cloud-aiplatform). Actively maintained—released 10 days ago with 5633 GitHub stars and no known vulnerabilities. Requires Python 3.11+.

License in practice

Apache-2.0 permissive license allows commercial and private use with minimal restrictions; suitable for most enterprise and open-source projects.

Quickstart

# Install via uv (recommended)
uvx google-agents-cli setup

# Or use directly
from google_agents_cli import cli

# Run a command
cli.main(['scaffold', 'my-agent'])

Requires Python 3.11+. For full functionality, Google Cloud credentials and optionally Node.js are needed; local development can use AI Studio API key.

Verify before relying

• Whether the package works standalone without a coding agent IDE installed, or if integration with Antigravity/Claude Code/Codex is required for practical use.
• Performance and resource overhead of the 15 runtime dependencies in typical agent development workflows.
• Whether all CLI commands (deploy, publish, eval) work offline or require active Google Cloud connectivity.

Package facts

Evidence: google_agents_cli-1.3.1-py3-none-any.whl

Tags

More Build Tools packages

Further reading