Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

engineering-devops-automator工程 DevOps 自动化

Agent Skill

用于辅助云资源、部署、容器、基础设施和运维自动化任务。它适合让 Agent 检查配置、整理部署步骤、分析资源状态、生成排障思路或辅助云服务接入。使用时需要明确目标环境、账号权限、区域和资源组,区分本地测试与生产操作;涉及删除资源、重启服务、修改网络或权限配置时,应先确认影响范围。

总安装

665

周安装

28

GitHub Stars

8

下载量

233
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:engineering-devops-automator(工程 DevOps 自动化)
来源仓库:https://github.com/peterhdd/agent-skills
仓库路径:skills/engineering-devops-automator
安装命令:
npx skills add https://github.com/peterhdd/agent-skills --skill engineering-devops-automator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/peterhdd/agent-skills --skill engineering-devops-automator

简介

DevOps 自动化用于基础设施和部署流程的自动化管理。

  • 适合云资源、CI/CD 和监控告警配置场景。
  • 提供基础设施即代码的最佳实践指导。
  • 使用时需区分测试和生产环境操作。engineering-devops-automator 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 涉及资源删除或网络变更时应先评估影响范围。

SKILL.md

DevOps & Infrastructure Guide

Overview

This guide covers infrastructure automation, CI/CD pipeline development, deployment strategies, monitoring, and cloud operations. Use it when provisioning infrastructure, building pipelines, setting up observability, managing secrets, or planning disaster recovery.

First 10 Minutes

  • Inventory the delivery surface before proposing changes: CI config, infrastructure directories, runtime manifests, Dockerfiles, and observability config.
  • Run the existing validation commands before editing anything. If the repo has no validation path for infra changes, add one as part of the task.
  • Use scripts/analyze_deployment_risk.py on the repo root to summarize CI, Docker, Terraform, and Kubernetes signals before proposing rollout changes.
  • Identify the rollback path for the current deploy system. If you cannot explain how to revert the change in under 5 minutes, the rollout plan is incomplete.

Refuse or Escalate

  • Refuse "just push it" requests when there is no rollback path, no health signal, or no way to test the change outside production.
  • Escalate before changing production state if the plan includes database replacement, Terraform destroys, state moves, certificate rotation, or security group broadening without a compensating control.
  • Escalate when the repo mixes multiple deployment systems and ownership boundaries are unclear. Untangling that is a separate task.
  • Do not recommend Kubernetes by default. If the workload is a single service with simple networking and predictable scale, stay with the simpler runtime.

Infrastructure Decision Rules

Provisioning

  • Use Terraform with remote state (S3 + DynamoDB lock) so every resource is version-controlled and safe from concurrent modifications.
  • Use Terraform workspaces or directory-per-environment layout with shared modules to catch drift between staging and production.
  • Use the same Terraform modules as production with variable overrides -- never create infrastructure via cloud console.

CI/CD Pipelines

  • Structure as discrete stages (lint, test, build, scan, deploy) with explicit dependencies so security failures block deployment.
  • Deployment strategy: blue-green for zero-downtime + instant rollback, canary for gradual traffic shifting with metric-based promotion, rolling when simplicity matters and brief mixed-version traffic is acceptable.
  • Automate any manual step performed more than twice; delete the manual runbook entry to prevent drift.

Containerization

  • Use multi-stage Docker builds with distroless or Alpine final images to minimize attack surface.
  • CI must run Trivy (or equivalent) and fail on CRITICAL/HIGH findings before merge.

Monitoring and Reliability

  • Instrument the four golden signals (latency, traffic, errors, saturation); alert on symptoms, not causes.
  • Every alert must link to a runbook; alerts without runbooks get deleted or converted to dashboard metrics within one sprint.
  • Enforce structured JSON logging; ship to centralized system (ELK, Loki) with compliance-aligned retention.
  • Configure liveness probes for 30-second restart; set PodDisruptionBudget for availability during disruptions.

Disaster Recovery

  • Automate failover with runbooks tested quarterly; an untested DR plan is no plan.

Cost Optimization

  • Review cloud utilization monthly; downsize any instance averaging below 20% CPU over 14 days.

Secrets Management

  • Store secrets in Vault or AWS Secrets Manager with automated rotation (max 90-day TTL); inject at runtime.
  • Never commit secrets to source control or bake them into images.

Network Security

  • Default all security groups and NACLs to deny-all inbound; open only required ports/CIDRs; prune monthly.

Compliance

  • Generate automated audit logs recording deployer, commit SHA, and approval; store immutably for retention period.

Incident Response Protocol

  • Severity 1 (site down, data loss risk): Assemble incident team within 5 minutes. First action: mitigate (rollback, failover, scale up), not diagnose. Communicate status to stakeholders within 15 minutes. Post-mortem within 48 hours.
  • Severity 2 (degraded performance, partial outage): On-call engineer responds within 15 minutes. Check: recent deploys (rollback if <1 hour old), infrastructure alerts (CPU, memory, disk), dependency health (downstream services, databases). Communicate status within 30 minutes.
  • Severity 3 (minor issue, workaround exists): Log the issue, create a ticket, fix in next sprint. No immediate response required.
  • Rollback decision: If the issue started after a deploy within the last 4 hours, rollback first, investigate second. If the issue is not correlated with a deploy, escalate to the relevant service team.
  • Communication template: "We are aware of [impact description]. [X users / Y% of traffic] are affected. We are [current action]. Next update in [time]."

Cost Estimation Formulas

  • Compute (EC2/GCE): monthly_cost = instance_hourly_rate * 730 * instance_count. Reserved instances save 30-60% for steady-state workloads (commit for 1 year).
  • Storage (S3/GCS): monthly_cost = storage_GB * $0.023 + requests * $0.0004 (GET) or $0.005 (PUT). Enable lifecycle policies: move to Infrequent Access after 30 days, Glacier after 90 days.
  • Database (RDS/Cloud SQL): monthly_cost = instance_hourly_rate * 730 + storage_GB * $0.115 + IOPS_provisioned * $0.10. Multi-AZ doubles the instance cost.
  • Data transfer: First 1GB/month free. $0.09/GB out to internet. Inter-AZ: $0.01/GB each direction. Cross-region: $0.02/GB. Data transfer is the hidden cost — monitor it.
  • Kubernetes (EKS/GKE): cluster_cost = control_plane ($73/month EKS) + node_instance_costs + data_transfer. Spot/preemptible nodes save 60-90% for fault-tolerant workloads.
  • Rule of thumb: If cloud bill >$5k/month, hire a FinOps review. If >$50k/month, automate cost anomaly detection with AWS Cost Anomaly Detection or similar.

Service Selection Decision Trees

  • Compute: Lambda/Cloud Functions if <15 min execution, <10GB memory, and request-driven. ECS/Cloud Run for containerized services with consistent traffic. EKS/GKE only if running >10 services with complex networking requirements.
  • Database: RDS PostgreSQL for <10TB relational. DynamoDB for key-value at >100k QPS. ElastiCache Redis for caching and session storage. Aurora if you need PostgreSQL compatibility with automatic multi-AZ failover.
  • Queue/Messaging: SQS for simple async jobs. SNS + SQS for fan-out. EventBridge for event routing with filtering rules. Kafka (MSK) only for streaming >10k msg/sec with replay.
  • Storage: S3 for objects. EFS for shared filesystem (NFS). EBS for block storage (database volumes). Choose storage class based on access frequency.
  • CDN: CloudFront for AWS-native. Cloudflare for multi-cloud or DDoS-heavy. Use CDN for all static assets and any API response cacheable for >5 seconds.

Self-Verification Protocol

After any infrastructure or pipeline change, verify:

  • Terraform: Run terraform plan and read every line of the diff. If the plan shows any destroy or replace on a production resource, stop and verify intent.
  • CI/CD pipeline: Trigger a full pipeline run on a non-production branch. Verify every stage passes. Check that security scan gates actually block on findings (deliberately introduce a known CVE to test).
  • Monitoring: After setting up alerts, trigger each alert manually (spike CPU, kill a health check, fill disk). Verify the alert fires within the expected time window and reaches the correct channel.
  • Disaster recovery: After configuring backups, perform a restore to a test environment. Verify data integrity. If you cannot restore, you do not have backups — you have a false sense of security.
  • Secrets: Verify no secrets appear in: CI/CD logs (mask variables), Docker image layers (docker history), Terraform state (use sensitive = true), or git history (git log -p | grep -i password).

Failure Recovery

  • Terraform state drift: Run terraform plan to see the drift. If drift is in a non-critical resource, run terraform apply to reconcile. If drift is in a critical resource (database, load balancer), investigate who/what changed it manually and reconcile carefully. Never blindly terraform apply when state shows unexpected changes.
  • CI/CD pipeline broken: Check the last successful run. Diff the pipeline config between last success and current failure. Common causes: expired secrets/tokens, dependency version bump, runner image update, or rate limiting from a registry.
  • Container OOM-killed in production: Check kubectl describe pod for the OOM event. Increase memory limits if under-provisioned. If memory usage grows linearly over time, the application has a memory leak — fix the app, not the limits.
  • Certificate expiry: Automate renewal with cert-manager (Kubernetes) or ACM (AWS). Set alerts for 30, 14, and 7 days before expiry. If expired: renew immediately, check all services using the cert, verify they pick up the new cert (may need pod restart).
  • Disk full: Identify what filled it: logs (rotate and compress), Docker images (prune unused), database WAL (check replication lag), or temp files. Fix the root cause; expanding the disk is a temporary measure.

Scripts

  • scripts/validate_dockerfile.sh -- Check a Dockerfile against common best practices: multi-stage builds, USER instruction, HEALTHCHECK, no latest tags, COPY over ADD, and.dockerignore presence. Run with --help for usage.
  • scripts/check_services.sh -- Check TCP connectivity and HTTP response for a list of host:port pairs. Reports status, latency, and HTTP status code. Run with --help for usage.

Code Examples

See CI/CD Pipeline Guide for a full GitHub Actions pipeline with security scanning, container build, and blue-green deployment with smoke tests.

See Infrastructure & Monitoring Guide for Terraform (launch template, ASG, ALB, CloudWatch alarm) and Prometheus configuration with alert rules.

Workflow

Step 1: Infrastructure Assessment

  • Audit existing infrastructure, deployment process, and monitoring gaps.
  • Map application dependencies and scaling requirements.
  • Identify security and compliance requirements for the target environment.

Step 2: Pipeline Design

  • Design CI/CD pipeline with security scanning integration.
  • Plan deployment strategy (blue-green, canary, rolling).
  • Create infrastructure as code templates.
  • Design monitoring and alerting strategy.

Step 3: Implementation

  • Set up CI/CD pipelines with automated testing.
  • Implement infrastructure as code with version control.
  • Configure monitoring, logging, and alerting systems.
  • Create disaster recovery and backup automation.

Step 4: Optimization and Maintenance

  • Monitor system performance and optimize resources.
  • Implement cost optimization strategies.
  • Create automated security scanning and compliance reporting.
  • Build self-healing systems with automated recovery.

Deliverables

  • Deployment strategy with explicit rollback steps, health gates, and ownership.
  • Infrastructure change summary listing stateful resources, blast radius, and approval points.
  • CI/CD plan covering lint, test, build, scan, deploy, and post-deploy verification.
  • Monitoring and alert checklist tied to the changed services, not a generic dashboard wishlist.

References

  • CI/CD Pipeline Guide -- GitHub Actions pipeline with security scanning, container build, and blue-green deployment.
  • Infrastructure & Monitoring Guide -- Terraform (launch template, ASG, ALB, CloudWatch alarm) and Prometheus configuration.
  • Kubernetes Patterns -- Production Deployment, HPA, PDB, ConfigMap/Secret mounting, Ingress with TLS, CronJob, and Helm values.
  • Docker Best Practices -- Multi-stage Dockerfiles (Node.js, Python, Go),.dockerignore, Docker Compose, and Trivy scanning.
  • Monitoring & Observability -- Structured logging, Prometheus metrics, Grafana dashboard, alert rules, OpenTelemetry tracing, and health checks.
  • Incident Triage -- Repo-first production incident flow, rollback decision tree, and evidence capture checklist.
  • Deployment Rollback Guide -- Canary, blue-green, rolling, schema-change, and feature-flag rollback patterns.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

37.17%
按下载量换算87

Claude

31.94%
按下载量换算74

Cursor

17.86%
按下载量换算42

Gemini CLI

10.88%
按下载量换算25

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills