Skill

role-architect:gcp-architect

Provides GCP architecture expertise for organization/project structure, VPC networking, data/analytics, Kubernetes/serverless, AI/ML platforms, security, and cost optimization. Use when designing or evaluating GCP systems.

GCP

Kubernetes

infrastructure

devops

npx claudepluginhub rnavarych/alpha-engineer --plugin role-architect

Popularity

Parent stars

Parent forks

Invocation

How this skill is triggered — by the user, by Claude, or both

Slash command

/role-architect:gcp-architect

User invocable

Model invocable

Inline context

Default effort

Tool Access

This skill is limited to the following tools:

ReadGrepGlobBash

Context Preview

The summary Claude sees in its skill listing — used to decide when to auto-load this skill

- Designing GCP organization, folder, and project structure for a new workload

Supporting Files

references/gcp-data-aiml-security-cost.mdreferences/gcp-platform-and-compute.md

SKILL.md

34 lines · ~596 tokens

Similar Skills

gcp-solution-architect

Designs GCP solutions aligned with the Cloud Architecture Framework for multi-service architecture, resource hierarchy, and production readiness reviews. Use for cross-domain tradeoffs and architecture review board prep.

3 files3 tools

vanguard-frontier-agentic

cloud-architecture

Design cloud infrastructure for AWS, GCP, or Azure. Plan compute, storage, networking, and compliance. Use when architecting cloud systems or migrating to cloud.

infrastructure-design

gcp-cloud

Provides GCP infrastructure patterns and best practices for Compute Engine, Cloud Functions, Cloud Storage, BigQuery, GKE. Includes Terraform examples, security tips, SQL patterns, and cost optimization.

devops-data

Stats

LanguageShell

Parent stars13

Parent forks1

MaintenanceGood

Last CommitMar 3, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

Stats

Actions

Help us improve

Share bugs, ideas, or general feedback.

GCP Architect

When to use

Designing GCP organization, folder, and project structure for a new workload

Selecting between GKE Autopilot, GKE Standard, and Cloud Run for container workloads

Architecting Shared VPC, global load balancing, or hybrid connectivity with Cloud Interconnect

Choosing between BigQuery, Spanner, AlloyDB, Firestore, and Bigtable for data storage

Designing Vertex AI training and serving pipelines or generative AI architectures

Configuring BeyondCorp, VPC Service Controls, or Security Command Center

Optimizing costs with Committed Use Discounts, Sustained Use Discounts, and FinOps practices

Core principles

Shared VPC by default — centralize network management, decentralize compute

Organization policies as guardrails — enforce at folder level before projects are created

Workload Identity everywhere — no service account keys on compute resources

Global load balancing — GCP's network is global; use it for traffic management

CUDs before Spot — commit to steady-state baseline first, then use Spot for burst

Reference Files

references/gcp-platform-and-compute.md — Architecture Framework five pillars, organization/project structure, Landing Zone, Shared VPC design, VPC Flow Logs, global load balancing, Cloud Armor, Cloud Interconnect, GKE Autopilot vs Standard, multi-cluster Fleet, Cloud Run, and Compute Engine MIGs

references/gcp-data-aiml-security-cost.md — BigQuery architecture patterns, Pub/Sub and Dataflow streaming pipeline, Cloud Spanner and AlloyDB, Firestore and Bigtable, Vertex AI training and serving, TPU architecture, generative AI with Vector Search; BeyondCorp and IAP, VPC Service Controls, Security Command Center, Cloud KMS and DLP; CUDs, Sustained Use Discounts, and FinOps with billing export to BigQuery

GCP Architect

When to use

Designing GCP organization, folder, and project structure for a new workload
Selecting between GKE Autopilot, GKE Standard, and Cloud Run for container workloads
Architecting Shared VPC, global load balancing, or hybrid connectivity with Cloud Interconnect
Choosing between BigQuery, Spanner, AlloyDB, Firestore, and Bigtable for data storage
Designing Vertex AI training and serving pipelines or generative AI architectures
Configuring BeyondCorp, VPC Service Controls, or Security Command Center
Optimizing costs with Committed Use Discounts, Sustained Use Discounts, and FinOps practices

Core principles

Shared VPC by default — centralize network management, decentralize compute
Organization policies as guardrails — enforce at folder level before projects are created
Workload Identity everywhere — no service account keys on compute resources
Global load balancing — GCP's network is global; use it for traffic management
CUDs before Spot — commit to steady-state baseline first, then use Spot for burst

Reference Files

references/gcp-platform-and-compute.md — Architecture Framework five pillars, organization/project structure, Landing Zone, Shared VPC design, VPC Flow Logs, global load balancing, Cloud Armor, Cloud Interconnect, GKE Autopilot vs Standard, multi-cluster Fleet, Cloud Run, and Compute Engine MIGs
references/gcp-data-aiml-security-cost.md — BigQuery architecture patterns, Pub/Sub and Dataflow streaming pipeline, Cloud Spanner and AlloyDB, Firestore and Bigtable, Vertex AI training and serving, TPU architecture, generative AI with Vector Search; BeyondCorp and IAP, VPC Service Controls, Security Command Center, Cloud KMS and DLP; CUDs, Sustained Use Discounts, and FinOps with billing export to BigQuery

role-architect:gcp-architect

Popularity

Invocation

Tool Access

Context Preview

Supporting Files

SKILL.md

Similar Skills

Help us improve

Help us improve

Find plugins for your project

role-architect:gcp-architect

Popularity

Invocation

Tool Access

Context Preview

Supporting Files

SKILL.md

GCP Architect

When to use

Core principles

Reference Files

Similar Skills

Help us improve

GCP Architect

When to use

Core principles

Reference Files