landing-zone-review
Skill1 starsChanged 3 months agoMITView on GitHub
Review a cloud landing zone against Cloud Adoption Framework design areas — management group and account topology, identity, network, security, governance, management and platform automation. Use when the user asks to review a landing zone, assess subscription/account structure, check CAF alignment, or evaluate the foundation of a cloud estate before workloads land on it.
In this repository4 other files it gives its agents.
- finops-reviewSkill
- terraform-security-reviewSkill
- threat-modelSkill
- well-architected-reviewSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 100
Do not pad the report. Five real findings beat twenty generic ones.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Landing Zone Review
Review the estate against the eight CAF design areas. Work from evidence — IaC, exported policy, tenant/account structure, network config. Where evidence is missing, say so; never assume a control exists.
Method
- Establish scope: which cloud(s), how many subscriptions/accounts, greenfield or brownfield.
- Gather evidence: Terraform/Bicep/CloudFormation, management-group or OU export, policy assignments, network topology, IAM roles, diagnostic settings.
- Assess each design area below.
- Rate each: Compliant / Partial / Gap / Not assessed. No rating without evidence.
- Rank findings by blast radius, not by design-area order.
A. Billing and tenant
- Single tenant or multi-tenant? Enterprise agreement / billing account structure clear?
- Are dev/test and production isolated at the billing boundary?
B. Identity and access management
- Central identity provider; SSO enforced; no local accounts on platform resources
- RBAC via groups, never direct user assignments
- Privileged access: just-in-time elevation, approval, time-bound (PIM / equivalent)
- Break-glass accounts: exist, excluded from conditional access, monitored, credentials escrowed
- Workload identity: managed identities / IRSA / workload identity federation — not static keys
- Separation of duties between platform and application teams
C. Resource organization
- Management group / OU hierarchy: shallow, purpose-built, not mirroring the org chart
- Landing zone separation: platform (identity, connectivity, management) vs application
- Subscription/account per environment and per blast-radius boundary
- Naming and tagging standard, enforced by policy not convention
- Mandatory tags: owner, cost centre, environment, data classification
D. Network topology and connectivity
- Hub-and-spoke or equivalent; transit strategy explicit
- Address space planned centrally, non-overlapping, room to grow
- Ingress and egress inspection points defined; forced tunnelling decision documented
- Private connectivity to PaaS (private endpoints / VPC endpoints) rather than public + firewall
- DNS: centralised resolution, private zones linked, split-horizon handled
- Segmentation between spokes: default-deny between landing zones
- DDoS protection on internet-facing entry points
E. Security
- Encryption at rest and in transit; key management ownership (platform vs workload) explicit
- Secrets in a vault; no secrets in IaC, pipelines or images
- Centralised security monitoring (SIEM) with platform + workload log sources onboarded
- Vulnerability and posture management enabled across all subscriptions/accounts
- Baseline hardening applied via policy; drift detected
- Data classification drives control selection, not the other way round
F. Management
- Central log destination; retention meets the longest legal/regulatory requirement
- Diagnostic settings applied by policy at scale, not per-resource
- Backup and recovery: defined RPO/RTO per tier, tested restores
- Patch and update management for platform components
- Monitoring covers platform health, not just workloads
G. Governance
- Policy-as-code, assigned at management group / OU scope, deny before audit where it matters
- Cost guardrails: budgets, alerts, and a named owner per subscription
- Compliance baseline mapped to the regulatory regime that actually applies
- Exemptions time-bound, justified, reviewed
H. Platform automation and DevOps
- Everything reproducible from IaC; console changes detected and reverted
- State/backends secured, versioned, locked
- Pipeline identities least-privileged and federated, not long-lived keys
- Environment promotion path defined; production requires approval
- Landing-zone changes versioned and reviewed like application code
Output
## Landing Zone Review — <estate>
| Design area | Rating | Key finding |
|---|---|---|
| Billing & tenant | | |
| Identity & access | | |
| Resource organization | | |
| Network & connectivity | | |
| Security | | |
| Management | | |
| Governance | | |
| Platform automation | | |
### Top findings by blast radius
1. <finding> — impact, evidence (file:line or resource), fix, effort
### Not assessed
<what you could not see, and what to provide>Do not pad the report. Five real findings beat twenty generic ones.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

