Google Agents CLI - Setup Guide¶
Source: google/agents-cli (357K+ combined installs) Category: Agent Infrastructure Quality Tier: 🟢 Production
Google's official Agent Development Kit (ADK) - a complete workflow for building, scaffolding, evaluating, deploying, and observing AI agents. Six skills covering the full agent lifecycle. Production-grade, maintained by Google.
Installation¶
# Install individual skills
npx skills add google/agents-cli --skill google-agents-cli-adk-code
npx skills add google/agents-cli --skill google-agents-cli-workflow
npx skills add google/agents-cli --skill google-agents-cli-eval
npx skills add google/agents-cli --skill google-agents-cli-scaffold
npx skills add google/agents-cli --skill google-agents-cli-deploy
npx skills add google/agents-cli --skill google-agents-cli-observability
Included Skills¶
| Skill | Installs | Purpose |
|---|---|---|
| google-agents-cli-adk-code | 59.8K | Core ADK coding patterns - build agents with Google's framework |
| google-agents-cli-workflow | 59.6K | Define multi-step agent workflows with chaining and branching |
| google-agents-cli-eval | 59.5K | Evaluate agent performance with structured test suites |
| google-agents-cli-scaffold | 59.5K | Scaffold new agent projects from templates |
| google-agents-cli-deploy | 59.5K | Deploy agents to Google Cloud / Vertex AI |
| google-agents-cli-observability | 59.5K | Monitor agent runs with traces, logs, and metrics |
Prerequisites¶
| Requirement | Details |
|---|---|
| Google Cloud account | Required for deploy and observability skills |
| Python 3.10+ | Core runtime |
| gcloud CLI | For deployment workflows |
Key Capabilities¶
Agent Development (adk-code)¶
Full Google ADK development patterns - tool definitions, memory management, multi-turn conversations, streaming responses, and safety filters.
Agent Workflows (workflow)¶
Chain agents together, define conditional branching, parallel execution, and human-in-the-loop checkpoints.
Agent Evaluation (eval)¶
Structured eval harness - test datasets, metric definitions, regression detection, and A/B comparison between agent versions.
Scaffolding (scaffold)¶
adk init equivalent - project templates, directory structures, config files, and CI/CD pipeline generation.
Deployment (deploy)¶
Push agents to Vertex AI Agent Builder, Cloud Run, or GKE. Environment variable management and canary rollout support.
Observability¶
OpenTelemetry-based tracing for agent runs, latency breakdowns by step, token usage tracking, and error rate dashboards.
Quick Start¶
# Scaffold a new agent project
npx skills use google/agents-cli@google-agents-cli-scaffold
# Build the agent logic
npx skills use google/agents-cli@google-agents-cli-adk-code
# Define multi-step workflows
npx skills use google/agents-cli@google-agents-cli-workflow
# Evaluate before deploying
npx skills use google/agents-cli@google-agents-cli-eval
# Deploy to Google Cloud
npx skills use google/agents-cli@google-agents-cli-deploy
# Monitor in production
npx skills use google/agents-cli@google-agents-cli-observability
Verification¶
npx skills list | grep google-agents-cli
Notes¶
- Official Google product - actively maintained, frequent updates
- All 6 skills have near-identical install counts (~59.5K), indicating they're typically installed as a bundle
- Requires Google Cloud for deploy/observability skills; code/eval/scaffold work standalone
- Complements Hermes agent infrastructure for production deployment patterns
- Quality tier 🟢 Production: 357K+ combined installs, Google-maintained
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.