cloud-infrastructure · ormus-solutions — плагин для Claude Code: обзор и установка
Плагин cloud-infrastructure проектирует облачную инфраструктуру под AWS, GCP и Azure, собирает модули Terraform и помогает снизить расходы на облака через шесть специализированных субагентов.
MIT ★ 126 обновлён 2026-10-04
/plugin marketplace add Ormus-Solutions/LibreUIUX-Claude-Code
/plugin install cloud-infrastructure@libreuiuxЧто делает cloud-infrastructure
Плагин добавляет в Claude Code набор практик по архитектуре облаков, сетям и Kubernetes. Здесь нет готовых слэш-команд. Всё взаимодействие строится на вызове субагентов через собачку, таких как @cloud-architect, @terraform-specialist или @kubernetes-architect.
Внутри находятся четыре базы знаний по модулям Terraform, мультиоблачным паттернам, гибридным сетям и сокращению затрат. Агенты генерируют код для создания сетей VPC, кластеров EKS, GKE и AKS, баз данных и политик жизненного цикла для объектных хранилищ S3 или GCS.
Поскольку плагин лишён исполняемых команд, обращаться к его возможностям нужно строго через имена агентов в тексте запроса.
Какие задачи решает
- Спроектировать отказоустойчивую архитектуру в AWS, GCP или Azure
- Подготовить повторно используемые модули Terraform для сетей и баз данных
- Связать локальный дата-центр с облачной инфраструктурой через VPN или Direct Connect
- Оптимизировать расходы на облако за счёт правил хранения и резервирования ресурсов
Примеры запросов агенту
- @cloud-architect спроектируй отказоустойчивую инфраструктуру в AWS с кластером EKS и базой RDS
- @terraform-specialist создай модуль VPC с тремя зонами доступности и NAT Gateway
- @hybrid-cloud-architect покажи, как выстроить сеть между локальным ЦОД и AWS через Direct Connect
Как установить cloud-infrastructure
Через маркетплейс плагинов
/plugin marketplace add Ormus-Solutions/LibreUIUX-Claude-Code/plugin install cloud-infrastructure@libreuiuxПо шагам, с командами для каждой платформы: Плагины Claude Code: маркетплейсы, установка и как собрать свой
Вопросы о cloud-infrastructure
Как установить cloud-infrastructure в Claude Code?
Выполните в терминале две команды по очереди: /plugin marketplace add Ormus-Solutions/LibreUIUX-Claude-Code, а затем /plugin install cloud-infrastructure@libreuiux.
Какие агенты входят в состав cloud-infrastructure?
В плагин включены шесть агентов: cloud-architect, terraform-specialist, network-engineer, kubernetes-architect, hybrid-cloud-architect и deployment-engineer.
Есть ли в плагине слэш-команды?
Нет, плагин не содержит слэш-команд и управляется только прямыми обращениями к агентам.
Исходный текст
cloud-infrastructure · MIT
# Cloud Infrastructure Plugin
> Multi-cloud architecture, Terraform modules, and infrastructure-as-code mastery.
**Cloud Infrastructure** provides:
- **4 Skills** - Terraform, multi-cloud, networking, cost optimization
- **6 Agents** - Cloud architects for AWS, GCP, Azure, hybrid, networking
- **No Commands** - Agent-driven workflows
**Result**: Scalable, cost-effective cloud infrastructure defined as code.
---
## Quick Start
### Invoke Agents
```
@cloud-architect design a multi-region AWS setup for high availability
@terraform-specialist create a reusable VPC module
@network-engineer design hybrid connectivity between AWS and on-prem
@kubernetes-architect plan a production EKS deployment
```
---
## The 4 Skills
### 1. Terraform Module Library
> Reusable, production-ready infrastructure modules
**Core Modules:**
| Module | Purpose |
|--------|---------|
| VPC | Network isolation, subnets, routing |
| EKS/GKE/AKS | Managed Kubernetes clusters |
| RDS/CloudSQL | Managed databases |
| S3/GCS | Object storage with policies |
| IAM | Identity and access management |
**Best Practices:**
```hcl
# Module structure
module "vpc" {
source = "./modules/vpc"
name = "production"
cidr = "10.0.0.0/16"
availability_zones = ["us-west-2a", "us-west-2b", "us-west-2c"]
private_subnets = ["10.0.1.0/24", "10.0.2.0/24", "10.0.3.0/24"]
public_subnets = ["10.0.101.0/24", "10.0.102.0/24", "10.0.103.0/24"]
enable_nat_gateway = true
single_nat_gateway = false # HA: one NAT per AZ
}
```
---
### 2. Multi-Cloud Architecture
> Design principles for AWS, GCP, and Azure
**Cloud Comparison:**
| Concern | AWS | GCP | Azure |
|---------|-----|-----|-------|
| Compute | EC2, ECS, EKS | Compute Engine, GKE | VMs, AKS |
| Database | RDS, DynamoDB | CloudSQL, Firestore | CosmosDB, SQL |
| Serverless | Lambda | Cloud Functions | Functions |
| Storage | S3 | GCS | Blob Storage |
**Multi-Cloud Patterns:**
- **Active-Active**: Workloads on multiple clouds
- **DR/Backup**: Primary on one, DR on another
- **Best-of-Breed**: Use each cloud's strengths
- **Avoid Lock-in**: Containerize, use Terraform
---
### 3. Hybrid Cloud Networking
> Connect cloud, on-premises, and edge
**Connectivity Options:**
| Option | Speed | Use Case |
|--------|-------|----------|
| Site-to-Site VPN | ~1.25 Gbps | Quick setup, backup |
| Direct Connect/Interconnect | 1-100 Gbps | Production workloads |
| Transit Gateway | Hub-and-spoke | Multi-VPC, multi-region |
| Mesh VPN | Varies | Multi-site |
**Architecture:**
```
On-Premises DC
|
Direct Connect
|
Transit Gateway ---- VPC Production
| ---- VPC Staging
| ---- VPC Development
|
Shared Services VPC
```
---
### 4. Cost Optimization
> Reduce cloud spend without sacrificing performance
**Key Strategies:**
| Strategy | Savings | Effort |
|----------|---------|--------|
| Reserved Instances | 30-60% | Low |
| Spot/Preemptible | 60-90% | Medium |
| Right-sizing | 20-40% | Medium |
| Auto-scaling | Variable | High |
| Storage tiering | 30-50% | Low |
**Quick Wins:**
```hcl
# Enable S3 lifecycle policies
resource "aws_s3_bucket_lifecycle_configuration" "example" {
bucket = aws_s3_bucket.example.id
rule {
id = "archive-old-objects"
status = "Enabled"
transition {
days = 90
storage_class = "GLACIER"
}
}
}
```
**Monitoring:**
- AWS Cost Explorer / GCP Billing / Azure Cost Management
- Set budget alerts
- Tag resources for cost allocation
- Review unused resources weekly
---
## The 6 Agents
| Agent | Expertise |
|-------|-----------|
| **cloud-architect** | High-level cloud strategy and design |
| **terraform-specialist** | IaC, modules, state management |
| **network-engineer** | VPCs, peering, hybrid connectivity |
| **kubernetes-architect** | Managed K8s, cluster design |
| **hybrid-cloud-architect** | Multi-cloud, on-prem integration |
| **deployment-engineer** | CI/CD, deployment automation |
---
## Plugin Structure
```
cloud-infrastructure/
├── README.md # This file
│
├── skills/ # Knowledge base
│ ├── terraform-module-library/
│ │ ├── SKILL.md # Module patterns
│ │ └── references/
│ │ └── aws-modules.md # AWS-specific modules
│ ├── multi-cloud-architecture/
│ │ └── SKILL.md # Multi-cloud patterns
│ ├── hybrid-cloud-networking/
│ │ └── SKILL.md # Networking patterns
│ └── cost-optimization/
│ └── SKILL.md # Cost patterns
│
└── agents/ # Specialist agents
├── cloud-architect.md # Strategy & design
├── terraform-specialist.md # IaC expert
├── network-engineer.md # Networking expert
├── kubernetes-architect.md # K8s expert
├── hybrid-cloud-architect.md # Hybrid/multi-cloud
└── deployment-engineer.md # Deployment automation
```
---
## Usage Examples
### Example 1: New AWS Environment
```
User: Set up a production AWS environment with EKS, RDS, and proper networking
@cloud-architect designs:
1. VPC Architecture
- 3 AZs for high availability
- Public subnets for ALB
- Private subnets for EKS/RDS
- NAT Gateways per AZ
2. EKS Cluster
- Managed node groups
- Cluster autoscaler
- ALB Ingress Controller
- External Secrets Operator
3. RDS PostgreSQL
- Multi-AZ deployment
- Automated backups
- Parameter groups optimized
4. Security
- Security groups (least privilege)
- IAM roles for service accounts
- Secrets in AWS Secrets Manager
```
### Example 2: Cost Reduction
```
User: Our AWS bill is $50k/month, help reduce it
@cloud-architect analyzes:
1. Compute (40% of bill)
- 20 instances never scale down -> Auto Scaling Groups
- Dev/staging run 24/7 -> Scheduled scaling
- No reserved capacity -> Purchase 1-year RIs
2. Storage (25% of bill)
- 5TB in S3 Standard (rarely accessed) -> Glacier
- EBS volumes oversized -> Right-size based on metrics
3. Networking (15% of bill)
- NAT Gateway data transfer -> Consider VPC endpoints
- Inter-region traffic -> Cache locally
Projected savings: $18k/month (36%)
```
### Example 3: Multi-Cloud Strategy
```
User: We need to avoid AWS lock-in while using their services
@hybrid-cloud-architect recommends:
1. Containerize Everything
- EKS now, but Kubernetes is portable
- Use Helm for deployments
2. Abstract Cloud Services
- Use Terraform for all clouds
- Store state in cloud-agnostic backend
3. Data Portability
- Standard formats (PostgreSQL, not DynamoDB)
- S3-compatible storage (works with GCS, MinIO)
4. Monitoring
- Cloud-agnostic: Datadog, Grafana Cloud
- Or self-hosted: Prometheus + Grafana
```
---
## When to Use This Plugin
### Perfect For:
- Cloud architecture design
- Infrastructure-as-code (Terraform)
- Multi-cloud and hybrid strategies
- Network design and connectivity
- Cost optimization initiatives
### Combines Well With:
- **kubernetes-operations** - K8s workload management
- **cicd-automation** - Infrastructure CI/CD
- **security-compliance** - Cloud security posture
- **observability-monitoring** - Cloud monitoring
---
## Technology Reference
| Category | Technologies |
|----------|--------------|
| IaC | Terraform, Pulumi, CloudFormation, CDK |
| Cloud Providers | AWS, GCP, Azure, DigitalOcean |
| Networking | VPC, Direct Connect, Transit Gateway |
| Container Platforms | EKS, GKE, AKS, OpenShift |
| Cost Tools | AWS Cost Explorer, Infracost, Spot.io |
---
## Key Patterns
### Terraform State Management
```hcl
terraform {
backend "s3" {
bucket = "company-terraform-state"
key = "production/vpc/terraform.tfstate"
region = "us-west-2"
encrypt = true
dynamodb_table = "terraform-locks"
}
}
```
### Environment Separation
```
environments/
├── production/
│ ├── main.tf
│ └── terraform.tfvars
├── staging/
│ ├── main.tf
│ └── terraform.tfvars
└── development/
├── main.tf
└── terraform.tfvars
modules/
├── vpc/
├── eks/
└── rds/
```
### Security Best Practices
- Encrypt everything at rest and in transit
- Use IAM roles, not access keys
- Enable CloudTrail/Audit logging
- Implement VPC flow logs
- Use private endpoints where possible
---
## License
Part of the LibreUIUX project - MIT License
---
*"Infrastructure should be as reliable as you forget it exists."*
© Seth Hobson · MIT · Ormus-Solutions/LibreUIUX-Claude-Code
Похожие плагины
Все плагины по теме «Облака: AWS, Azure, GCP» →Сначала — cloud-infrastructure от других авторов.