Automation & Ops ✓ Google · Official google-cloudai-agent-architecturemulti-agent-systemsinfrastructure-as-codeterraformcloud-runmodel-context-protocoldeployment-validation

Google Cloud AI Agent Builder

Design, implement, deploy, and validate AI agents and multi-agent systems on Google Cloud.

FollowSkills review · FSRS-2.0
Not recommended
53/ 100 5-point scale 2.7 / 5
Trust16 / 25 · 3.2/5

The workflow requires confirmation of requirements, architecture, implementation, and validation, and explicitly mentions least privilege, OIDC, Model Armor, human oversight, encrypted environment variables, and deployment dry runs. However, it lacks concrete IAM boundaries, complete data-flow and sensitive-data lifecycle guidance, and an actionable rollback procedure; IaC and deployment can also cause external changes, so points were deducted.

Reliability8 / 20 · 2.0/5

The four-phase workflow is internally coherent and includes templates, validation checks, and deployment troubleshooting assistance. However, there are no executed tests, pinned dependency versions, complete reproducible command examples, or specific failure-diagnosis and recovery procedures. Much of the workflow depends on external references and placeholders, so the static score remains below 10.

Adaptability10 / 15 · 3.3/5

The name, target scenario, and exclusions for general architecture or narrow single-product tasks are reasonably clear, and the skill selects between single-agent and multi-agent patterns based on requirements. Inputs, output contracts, and trigger conditions remain broad; Chinese-language interaction is not addressed, and mainland-China reachability of Google Cloud services is not discussed, so points were deducted.

Convention9 / 15 · 3.0/5

SKILL.md has a clear phased structure and supplies guidance references, product mappings, design principles, and three output templates. Repository-level installation, Apache-2.0 licensing, contribution channels, and active-development status are documented. The skill lacks its own version, changelog, named maintenance responsibility, pinned dependencies, FAQ, and complete troubleshooting guidance, so points were deducted.

Effectiveness6 / 15 · 2.0/5

The skill clearly targets usable deliverables including an architecture document, Mermaid diagram, IaC, deployment instructions, and a validation plan. However, the templates are mostly skeletons, exact ADK/MCP code and deployment details must be generated dynamically from external material, and no representative output is verified. The score is therefore conservatively limited by static calibration.

Verifiability4 / 10 · 2.0/5

Evidence includes the skill's reference inventory, product mappings, templates, and repository-level license and installation documentation, primarily pointing to official material. There is no supplied test suite, CI coverage, execution log, or independent reproduction evidence, so key paths cannot be statically verified and the score is kept at or below 5.

Evidence confidence:Low Reviewed Jul 28, 2026 Reviewed revision d1c9be2009ba
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • IaC and deployment may modify real cloud resources; define the project, region, IAM scope, approval authority, change boundary, and rollback plan before execution.
  • Do not treat template placeholders, sample commands, or product recommendations as verified configurations; check current official documentation, API availability, quotas, regional support, and costs.
  • Chinese-language support and mainland-China network reachability are not specified; validate access conditions for the relevant Google Cloud, documentation, and CLI services.
  • Add workload-specific rules for sensitive-data inspection, log redaction, secret management, and agent-tool permissions.
See the full review method →

What does this skill do, and when should you use it?

This skill is for users designing or implementing AI agent systems on Google Cloud. It structures the work into requirements discovery, solution design, implementation planning, and validation. It recommends single-agent or multi-agent patterns based on workload complexity, workflow, latency, and cost, then produces architecture, deployment, and verification materials. Its scope is agentic Google Cloud solutions, not general cloud architecture or narrow single-product tasks.

Collects functional and non-functional requirements, current systems, knowledge sources, and dependencies; identifies ambiguities and requests confirmation; recommends a single-agent or multi-agent pattern; decomposes components and maps them to Google Cloud products; creates Mermaid architecture diagrams; and drafts solution-architecture.md, implementation-instructions.md, and validation-plan.md. The planned materials can include Terraform, Agents CLI, gcloud, curl, Cloud Run scaling guidance, and MCP-based tool registration.

  1. A cloud architect needs to discover requirements and confirm a technical decomposition for a new AI-agent workload.
  2. A platform engineer needs Google Cloud product mappings, an architecture diagram, and deployment guidance for an agent system.
  3. A development team needs implementation instructions containing Terraform or Agents CLI steps.
  4. An operations team needs checks for routing, IAM enforcement, security policies, and deployed service endpoints.
  5. A team needs local agent runs, evaluations, and deployment dry runs before production release.

What are this skill's strengths and limitations?

Pros
  • Covers requirements, architecture, implementation planning, and validation in one workflow.
  • Requires confirmation at important design and review stages and supports iteration.
  • Specifies Mermaid diagrams, Terraform, Agents CLI, MCP, Cloud Run cost controls, and protected credential handling.
  • Accounts for cross-cloud, hybrid, and on-premises components.
Limitations
  • Its full document-generation workflow depends on referenced guidance and template files whose contents are not provided here.
  • It does not supply a ready-made architecture for a workload; sufficient requirements and user confirmation are required.
  • The source provides no test suite, platform support matrix, or evidence of an actual deployment.
  • Implementation and validation depend on Google Agents CLI, Terraform, gcloud CLI, curl, and suitable cloud permissions.

How do you install this skill?

Install the collection with npx skills add google/skills. During installation, select skills/cloud/google-cloud-solution-build-deploy-agents. The source does not document a specific installation directory or separate skill-only command.

How do you use this skill?

In an Agent Skills-compatible client, provide a concrete request such as: Design, implement, and validate a multi-agent system on Google Cloud that accesses an enterprise knowledge base. Include functional requirements, non-functional requirements, existing systems, knowledge sources, and dependencies; the skill requests confirmation at successive phases.

How does this skill compare with similar options?

Unlike the repository's Google Cloud solution-architecture skill, this skill is specifically for AI-agent or multi-agent systems; general Google Cloud solution architecture is directed to that other skill.

FAQ

Does it support single-agent systems?
Yes. It recommends either a single-agent or multi-agent pattern based on complexity, workflow, latency, and cost requirements.
Does it deploy resources automatically?
The source says it generates IaC, scripts, and deployment instructions and may use Agents CLI where appropriate; it does not establish that it performs deployment on the user's behalf.
Is MCP required?
The implementation instructions must demonstrate registering tools such as database readers using MCP standards, so MCP configuration is required for those tool-integration scenarios.
Can it replace a general Google Cloud architecture skill?
No. It is explicitly limited to designs and implementations with agent context, rather than general Google Cloud architecture or narrow single-product work.

More skills from this repository

All from google/skills

Automation & Ops ✓ Google · Official

Google Cloud Live Multimodal Streaming Architect

Design and deploy Google Cloud solutions for live, bidirectional multimodal streams.

Automation & Ops ✓ Google · Official

Google Cloud Secure Serverless Tiers

Guides agents to design, codify, deploy, and validate isolated multi-tier serverless web applications on Google Cloud.

Automation & Ops ✓ Google · Official

Google Cloud Solution Architect

Plan, validate, and package end-to-end architectures for complex multi-product Google Cloud workloads.

Automation & Ops ✓ Google · Official

Google Cloud Operational Excellence Advisor

Assesses Google Cloud workloads and recommends improvements using the WAF Operational Excellence pillar.

Dev & Engineering ✓ Google · Official

Google Agents CLI Agent Development Guide

Guides agents from specification through development, deployment, and monitoring.

Automation & Ops ✓ Google · Official

Vertex AI Agent Alerting

Configure dynamic reliability and quality alerts for Agent Runtime agents.

Automation & Ops ✓ Google · Official

GKE Enterprise RAG Search Architect

Designs and validates enterprise RAG search systems built on GKE and AlloyDB.

Data & Analysis ✓ Google · Official

AlloyDB Database Manager

Manage AlloyDB clusters, instances, and backups with CLI, IaC, client-library, and MCP guidance.

Automation & Ops ✓ Google · Official

Google Cloud Global Load Balancer Configurator

Design and deploy opinionated global external Application Load Balancers on Google Cloud.

Data & Analysis ✓ Google · Official

Google Cloud Borderless Lakehouse Architect

Designs governed lakehouses that connect multicloud data to AI agents.

Data & Analysis ✓ Google · Official

Google Cloud Agent Evaluation Flywheel

Improve model and agent quality through structured evaluation, failure analysis, and iteration.

Automation & Ops ✓ Google · Official

Agent Platform Model Deployment Assistant

Deploy, verify, and clean up Model Garden models on Google Agent Platform.

Automation & Ops ✓ Google · Official

AI Workload Migration to GKE Inference

Move existing AI inference workloads to self-hosted inference on Google Kubernetes Engine.

Dev & Engineering ✓ Google · Official

Gemini API Enterprise Development Guide

Helps developers build enterprise Gemini applications on Agent Platform with Google's unified Gen AI SDK.

Data & Analysis ✓ Google · Official

Agent Platform Model Registry Manager

Manage Agent Platform models, versions, and metadata from the gcloud CLI.

Dev & Engineering ✓ Google · Official

Google Cloud Agent Platform Inference Assistant

Guides developers in connecting to Google Cloud Agent Platform and calling Gemini or OpenMaaS models.

Data & Analysis ✓ Google · Official

Google Cloud Agentic Data Science Architect

Design and validate a multi-product Google Cloud architecture for agent-based data analytics and machine learning workloads.

Data & Analysis ✓ Google · Official

Agent Platform Model Tuning

A guided end-to-end workflow for fine-tuning open or Gemini models on Google Agent Platform.

Automation & Ops ✓ Google · Official

Google Cloud Storage Basics

Manage GCS buckets and objects while configuring access, protection, lifecycle, cost, and performance.

Automation & Ops ✓ Google · Official

Cloud Run Deployment Guide

Deploy and manage Cloud Run services, jobs, and worker pools with gcloud.

Related skills