What Kind of Skills Are Included in the Google Skills Repository? A Complete Category Guide

The Google Skills repository contains agent-skills organized into 12 categories: Google Cloud onboarding, multi-product solutions, AI/ML, infrastructure, databases & analytics, developer tools, management tools, Well-Architected Framework, security & identity, web & app hosting, advertising, and external integrations.

The Google Skills repository is a modular collection of agent-skills designed to enable AI agents to interact with Google products and services. These skills are consumed by the Agent Platform and other agent harnesses to perform concrete tasks ranging from GKE cluster provisioning to Google Ads campaign management. Each skill is defined in a SKILL.md file with optional reference implementations in Python, Java, Go, or other languages.

Overview of Skill Categories

According to the Google Skills source code, the repository organizes skills into thematic directories under skills/. Below is the complete breakdown of what types of skills are included, with representative examples from each category.

Getting Started with Google Cloud

Introductory skills for authentication and foundational onboarding. These enable agents to establish credentials before executing other operations.

• Example: skills/cloud/google-cloud-recipe-auth/SKILL.md — authenticates the agent to Google Cloud using service accounts or user credentials.

# Install the authentication skill

npx skills add google/skills/cloud/google-cloud-recipe-auth

Multi-Product Solution Skills

End-to-end workflows that orchestrate multiple Google Cloud services into cohesive solutions. These skills handle cross-service dependencies and deployment pipelines.

• Example: skills/cloud/google-cloud-solution-build-deploy-agents/SKILL.md — builds and deploys AI agents on Google Cloud infrastructure.

AI/ML Skills

Skills exposing machine learning capabilities through the Agent Platform, Gemini, BigQuery ML, and model registry operations.

• Example: skills/cloud/agent-platform-model-registry/SKILL.md — manages model registration, versioning, and deployment through the Agent Platform Model Registry.

from agents import Agent

agent = Agent(skills=[
    "google/skills/cloud/agent-platform-model-registry"
])

result = agent.run("List all models in the production registry")

Infrastructure Skills

Core Google Cloud provisioning workflows including GKE cluster creation, autoscaling configuration, and networking setup.

• Example: skills/cloud/gke-cluster-creation/SKILL.md — provisions GKE clusters with configurable node pools and networking.

Databases & Analytics Skills

Data platform operations spanning AlloyDB, BigQuery, Bigtable, Cloud SQL, Spanner, and lineage tooling.

• Example: skills/cloud/bigquery-basics/SKILL.md — executes queries, manages datasets, and handles BigQuery job operations.

Developer Tools

Local development helpers including gcloud CLI wrappers, Google Agents CLI integrations, and device platform utilities.

• Example: skills/cloud/developer-device-platform-basics/SKILL.md — configures local development environments for Google agent development.

Management Tools

Observability and cost management through Cloud Monitoring, Cloud Logging, GKE cost analysis, and SLO alerting.

• Example: skills/cloud/cloud-monitoring-chart-generation/SKILL.md — generates visualization charts from Cloud Monitoring metrics.

# Generate a CPU utilization chart for a GKE cluster

skills run cloud/cloud-monitoring-chart-generation \
    --metric=container.googleapis.com/container/cpu/utilization \
    --duration=1h \
    --output=cpu_chart.png

The skill executes helper scripts in cloud-monitoring-chart-generation/scripts/validate_chart.py to produce the final visualization.

Well-Architected Framework Skills

Guided implementations of the five pillars of the Google Cloud Well-Architected Framework: reliability, security, cost optimization, operational excellence, and performance efficiency.

• Example: skills/cloud/google-cloud-waf-cost-optimization/SKILL.md — applies cost-optimization patterns to deployed resources.

Security & Identity Skills

GKE platform security hardening, workload security policies, and detection-coverage evaluation.

• Example: skills/cloud/gke-platform-security/SKILL.md — implements security controls for GKE clusters and containerized workloads.

Web & App Hosting Skills

Serverless deployment and static hosting through Cloud Run and Firebase.

• Example: skills/cloud/cloud-run-basics/SKILL.md — deploys containerized services to Cloud Run.

Advertising Skills

Integration with Google Mobile Ads SDK, IMA SDK (both client-side and DAI), and Google Ads API automation.

• Example: skills/ads/google-mobile-ads-banner/SKILL.md — implements banner ad placements using the Mobile Ads SDK.

External & Other Skills

Links and references to related Google skill repositories including Google Analytics, ADK (Agent Development Kit), and Flutter.

• Example: skills/analytics/google-analytics-data-api-basics/SKILL.md — queries Google Analytics 4 data through the Data API.

Repository Structure and Skill Discovery

The Google Skills repository uses a consistent file layout that enables automatic discovery and execution:

• Top-level directories: skills/cloud/, skills/ads/, skills/analytics/ — product domains
• Skill definition: Each skill contains a SKILL.md with description, parameters, and usage instructions
• Reference implementations: Optional references/ subdirectory with language-specific code (reference/python.md, reference/java.md, etc.)
• Executable helpers: scripts/ folders with validation and utility code

The "Available Skills" section in the repository's README.md (lines 22-46) serves as the canonical registry used by the skill-discovery mechanism.

Summary

• 12 skill categories covering the full Google Cloud and advertising ecosystem
• Modular architecture: each skill is self-contained with SKILL.md definition and optional reference code
• Agent Platform integration: skills are consumed via npx skills add or Python Agent class instantiation
• Concrete file locations: skills live under skills/{domain}/{skill-name}/SKILL.md with predictable structure

Frequently Asked Questions

What is the Google Skills repository used for?

The Google Skills repository provides reusable agent-skills that enable AI agents to perform specific tasks with Google products. These skills are consumed by the Agent Platform and other agent frameworks to authenticate, provision resources, execute queries, and manage Google Cloud services programmatically.

How do I add a skill to my agent?

Install skills using the skills.sh CLI or Python SDK. For CLI installation, run npx skills add google/skills/cloud/{skill-name}. For Python agents, import the Agent class and pass skill identifiers to the skills parameter during initialization.

Where are individual skills defined in the repository?

Each skill is defined in a SKILL.md file located at skills/{category}/{skill-name}/SKILL.md. The repository's README.md contains a complete index under the "Available Skills" section, linking to all defined skills with descriptions and example use cases.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}