Skip to content
Back to skills

Cloud Core

ASecurity

Baseline multi-cloud primitives across AWS, Azure, and GCP: IAM least-privilege policies, VPC/VNet networking, secure compute, and CLI execution guardrails. Use when designing, reviewing, or implementing multi-cloud infrastructure primitives for: (1) IAM least-privilege policies and role design across AWS/Azure/GCP, (2) VPC, VNet, and Cloud VPC network topologies, (3) Secure compute instances and security groups/firewalls, or (4) CLI execution guardrails and dry-run safety.

  • 2 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 29, 2026
ai-agentsrailsawsgcpazuredatabasesecurity

Works with

  • cli

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 29, 2026

npx -y skills add talkops-ai/opscode --skill cloud-core --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cloud Core?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Cloud Core
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/talkops-ai-cloud-core/badge)](https://www.skillsdirectory.com/skills/talkops-ai-cloud-core)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: cloud-core
description: "Baseline multi-cloud primitives across AWS, Azure, and GCP: IAM least-privilege policies, VPC/VNet networking, secure compute, and CLI execution guardrails. Use when designing, reviewing, or implementing multi-cloud infrastructure primitives for: (1) IAM least-privilege policies and role design across AWS/Azure/GCP, (2) VPC, VNet, and Cloud VPC network topologies, (3) Secure compute instances and security groups/firewalls, or (4) CLI execution guardrails and dry-run safety."
license: MIT
compatibility: designed for opscode
---

# Cloud Core (Multi-Cloud Primitives & Guardrails)

Standard baseline patterns and safety guardrails across AWS, Azure, and GCP for identity, networking, compute, and execution safety.

## Quick Workflow

1. **Identify the Cloud Provider**: Determine whether the target task targets AWS, Azure, or GCP.
2. **Apply Identity Least-Privilege**: Design or review IAM roles/policies without wildcards (`*`) using scoped resources and condition keys.
3. **Establish Network Topology**: Implement multi-tier subnets (public/private/database) with explicit ingress/egress boundaries.
4. **Harden Compute Security**: Enforce workload identity / instance profiles, disable public IPs where possible, and use IMDSv2 / Shielded VMs.
5. **Run CLI Pre-flight Guardrails**: Validate target identity and run dry-run simulations before executing mutating commands.

---

## Provider Reference Manuals

Read the provider-specific guide corresponding to the target cloud platform:

- **AWS Primitives**: See [references/aws.md](references/aws.md) for IAM least-privilege policy templates, 3-AZ VPC topology, IMDSv2 compute hardening, and `aws` CLI dry-run flags.
- **Azure Primitives**: See [references/azure.md](references/azure.md) for custom RBAC roles, VNet/NSG subnet delegation, Managed Identities, and `az` CLI `--what-if` previews.
- **GCP Primitives**: See [references/gcp.md](references/gcp.md) for GCP IAM custom roles, Custom Mode VPC with Cloud NAT, Workload Identity, and `gcloud` CLI guardrails.
- **CLI Guardrails & Pre-Flight Checks**: See [references/guardrails.md](references/guardrails.md) for multi-cloud command safety matrices, identity verification, and execution checklists.

---

## Universal Multi-Cloud Guardrail Checklist

Execute these verification steps whenever performing infra modifications or executing cloud CLI commands:

- [ ] **Identity Alignment**: Verify current cloud credentials point to the intended target environment (account/subscription/project).
- [ ] **Least Privilege Scope**: Ensure IAM roles and policy statements do not contain unconstrained wildcard actions or resources.
- [ ] **Private Connectivity**: Confirm compute resources run in private subnets with egress via NAT gateways/Cloud NAT and ingress restricted by security groups or NSGs.
- [ ] **Dry-Run Validation**: Run `--dry-run`, `--what-if`, or plan previews prior to executing mutating or destructive operations.

Files in this skill

  • SKILL.md2.9 KB
  • references/aws.md3.3 KB
  • references/azure.md2.9 KB
  • references/gcp.md2.5 KB
  • references/guardrails.md1.9 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.