跳到主要内容
知仓学习社ZHICANG

operate-devops

Plan and implement infrastructure, CI/CD, container, deployment, observability, and operational configuration changes with least privilege, staged v…

不碰外部(只输出文字)无严重或高危命中hashgraph-online/awesome-codex-plugins

它会碰到什么

扫了多少3 个文本文件,5 KB
它会碰到什么不碰外部(只输出文字)
命中总数0 处
命中统计严重 0 · 高 0 · 中 0 · 低 0

这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。

技能内容

Operate DevOps

Workflow

  1. Read repository and environment guidance before changing configuration.
  2. Identify the target environment, current state, owning tool, blast radius, and approval boundary.
  3. Separate diagnosis, proposed change, apply, verification, and rollback.
  4. Prefer a dry run, plan, diff, render, lint, or local build before any apply step.
  5. Make the smallest reversible change.
  6. Validate configuration syntax and policy.
  7. Test failure behavior and rollback in the safest available environment.
  8. Review the diff for secrets, broad permissions, unpinned behavior, and unintended targets.

Guardrails

  • Confirm account, cluster, subscription, region, namespace, and environment before state-changing commands.
  • Keep credentials out of code, command output, and generated artifacts.
  • Use least-privilege identities and scoped approvals.
  • Do not silently apply infrastructure because a plan succeeded.
  • Do not delete, force-replace, rotate, deploy, or migrate live resources without explicit authorization.
  • Preserve a recovery path for availability, routing, storage, and identity changes.
  • Do not weaken security controls merely to make automation pass.
  • Treat third-party actions, base images, and pipeline dependencies as supply-chain inputs.

Change safety

Read [references/change-safety.md](references/change-safety.md) for deployment, infrastructure, CI, secret, permission, or data-path changes.

Verification

Use repository-native commands in this order:

format/lint → validate/render → plan/diff → policy tests → staged rollout check

Report exactly which environment was inspected and which commands were not run.

Acceptance criteria

  • Target and blast radius are explicit.
  • A preview or equivalent validation was inspected.
  • Credentials and permissions follow least privilege.
  • Rollback or roll-forward steps are concrete.
  • Monitoring proves success and detects regression.
  • The final report separates proposed, executed, and unverified actions.

想直接用这个技能?

本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。