DevOps & Cloud Kit

Plan-first infrastructure work: Terraform, AWS, containers, CI/CD pipelines, safe deployment and observability, with nothing applied or deployed until you approve it.

A setup for AI coding agents, made by Kitloom. One source, generated for every major agent: Claude Code, OpenAI Codex, Cursor, Google Antigravity, Gemini CLI, GitHub Copilot, Windsurf (Devin), opencode and Aider.

Install

1. Download this repository as a ZIP, or use the Check and download button on its Kitloom page, which checks it first.
2. Read the files. Nothing here runs on its own: the kit has no hooks and no install scripts.
3. Copy the folders and files you want into the root of your project. AGENTS.md is the shared instructions file; edit it to fit your project.

What is inside

Skills

• terrashark — Writing or reviewing Terraform/OpenTofu; catches identity churn, secret exposure and blast-radius mistakes.
• terraform-module-library — Designing reusable Terraform modules and standardising how cloud resources are provisioned.
• iam — Writing IAM policies and roles, or debugging AWS permission errors; keep access least-privilege.
• cloudwatch — Setting up AWS logs, metrics, alarms and dashboards, or querying logs with Insights.
• ln-92-container-engineer — Building, hardening or reviewing Dockerfiles and container images.
• github-actions-templates — Creating or editing GitHub Actions workflows for test, build and release.
• deployment-pipeline-design — Designing multi-stage pipelines with approval gates, canary or blue-green rollouts.
• secrets-management — Handling credentials in CI/CD and cloud: storage, injection and rotation.
• vibe-rollback-plan — Before any risky change: write down the rollback steps and blast radius first.
• vibe-safe-deploy — Preparing a deployment checklist with pre-flight checks and post-deploy verification, for the user to run or approve.
• observability-and-instrumentation — Adding logging, metrics, tracing and alerts so production behaviour is visible.
• cost-optimization — Reviewing cloud spend, rightsizing and tagging before or after a change.

Commands

Invoke by name (for example /plan-review).

• plan-review — Read-only review of a Terraform plan or infrastructure change: summarise creates, changes and destroys and flag risk.
• deploy-checklist — Produce a pre-deploy checklist with verification steps and a rollback plan, then wait for the user's decision.
• cost-check — Review infrastructure code and describe likely cost drivers and cheaper alternatives, without changing anything.

Sub-agents

• infra-reviewer — Delegate a read-only review of Terraform, Dockerfile and pipeline changes for safety, blast radius and security.
• incident-triager — Delegate read-only diagnosis of a failing deploy, pipeline or service from logs and config; it reports findings and proposes fixes.

Scoped rules

• terraform-safety — Applies when editing Terraform or OpenTofu files: format, validate, plan, and never apply.
• pipeline-changes — Applies when editing CI/CD workflow files: pin versions, least-privilege tokens, gated deploys.
• container-hygiene — Applies when editing Dockerfiles and compose files: small, pinned, non-root images.

Where each agent finds it

Agent	Instructions	Skills and commands	Rules	Sub-agents
Claude Code	CLAUDE.md (imports AGENTS.md)	.claude/skills/	.claude/rules/	.claude/agents/
OpenAI Codex	AGENTS.md	.agents/skills/	listed in AGENTS.md	.codex/agents/
Cursor	AGENTS.md	.agents/skills/, .claude/skills/	.cursor/rules/	.claude/agents/
Google Antigravity	AGENTS.md	.agents/skills/	.agents/rules/	.agents/agents/
Gemini CLI	AGENTS.md via .gemini/settings.json	.agents/skills/, .gemini/commands/	listed in AGENTS.md	.gemini/agents/
GitHub Copilot	AGENTS.md	.agents/skills/, .github/prompts/	.github/instructions/	.claude/agents/
Windsurf (Devin)	AGENTS.md	.agents/skills/, .windsurf/workflows/	.windsurf/rules/	—
opencode	AGENTS.md	.agents/skills/, .opencode/commands/	listed in AGENTS.md	.opencode/agents/
Aider	AGENTS.md via .aider.conf.yml	—	—	—

Skills are stored twice (.agents/skills/ and .claude/skills/) because Claude Code reads only its own folder. Agents that read both may list a skill twice; delete the copy you do not need.

Credits

The skills come from these open-source repositories, each under its own licence. Their licence and a notice are kept beside every copied skill.

Skill	Source	Licence
terrashark	LukasNiessen/terrashark at 21f2c18	MIT
terraform-module-library	wshobson/agents at 46891e7	MIT
iam	itsmostafa/aws-agent-skills at e786d25	MIT
cloudwatch	itsmostafa/aws-agent-skills at e786d25	MIT
ln-92-container-engineer	levnikolaevich/claude-code-skills at 161af7f	MIT
github-actions-templates	wshobson/agents at 46891e7	MIT
deployment-pipeline-design	wshobson/agents at 46891e7	MIT
secrets-management	wshobson/agents at 46891e7	MIT
vibe-rollback-plan	ash1794/vibe-engineering at 8f1d71b	MIT
vibe-safe-deploy	ash1794/vibe-engineering at 8f1d71b	MIT
observability-and-instrumentation	addyosmani/agent-skills at aeac41d	MIT
cost-optimization	wshobson/agents at 46891e7	MIT

Everything else in this repository (instructions, rules, commands, sub-agents) is by Kitloom under the MIT licence.

Use at your own risk

Read every file before you use it. Agent instructions can make an agent run commands on your machine. This kit was checked with Kitloom's pre-download check, which looks for warning signs but cannot prove anything is safe.