跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
部署与云服务

infrastructure

Workflow for provisioning cloud infrastructure using IaC. Use when the user needs to create or modify cloud resources.

girijashankarj · 已有 AI 审核记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add girijashankarj/cursor-handbook --skill infrastructure

技能说明


name: infrastructure description: Workflow for provisioning cloud infrastructure using IaC. Use when the user needs to create or modify cloud resources.

Skill: Set Up Cloud Infrastructure

Trigger

When the user needs to create or modify cloud resources.

Steps

Step 1: Define Requirements

  • What resources are needed?
  • Which environment? (dev, staging, production)
  • Performance requirements?
  • Budget constraints?
  • Security requirements?

Step 2: Design Architecture

  • Draw architecture diagram (Mermaid)
  • Define networking (VPC, subnets, security groups)
  • Plan compute resources (size, scaling)
  • Plan storage (database, file storage, cache)
  • Plan monitoring and alerting

Step 3: Write IaC

  • Create Terraform/CDK/CloudFormation templates
  • Use modules for reusable components
  • Parameterize environment-specific values
  • Follow naming conventions: {project}-{env}-{resource}
  • Tag all resources

Step 4: Security

  • Enable encryption at rest
  • Enable encryption in transit
  • Configure IAM with least privilege
  • Set up VPC and security groups
  • Enable audit logging

Step 5: Deploy

  • Plan: Review changes before applying
  • Apply to dev first
  • Verify in dev
  • Promote to staging
  • Verify in staging
  • Promote to production

Step 6: Monitor

  • Set up health checks
  • Configure alert rules
  • Create dashboards
  • Set up cost alerts

If a step fails

StepFailureRecovery
Step 5Deploy to dev/staging failsRun terraform destroy or equivalent; fix IaC; re-apply; never proceed to prod until dev/staging succeeds
Step 5Deploy to production failsRollback via IaC (revert and apply previous state); investigate; fix in dev; re-test full pipeline
Step 5Partial apply (some resources created)Document what was created; run destroy/rollback for failed resources; fix and retry

Never apply infrastructure changes to production without explicit user confirmation. Require confirmation before terraform apply or equivalent on prod.

Completion

Infrastructure provisioned, secured, monitored, and documented.