Hands-on labs I built to practice cloud security engineering end to end: writing controls as code, proving they work, and capturing evidence an assessor could use.
Every commit in this repo is signed, and every lab has an evidence/ folder with the raw output that shows the control working.
| Lab | What it proves | NIST 800-53 |
|---|---|---|
| 01 Kubernetes policy | Admission control and continuous audit with OPA Gatekeeper | AC-6, CM-2, CA-7 |
| 02 AWS detection | Config drift detection, GuardDuty, and CloudTrail alarms | AU-6, CA-7, SI-4, CM-3 |
| 03 Vulnerability scanning | Image and IaC scanning, SBOMs, and a CI gate that blocks critical CVEs | RA-5, SI-2, SA-11, CM-8 |
| 04 IAM least privilege | Scoped role assumption with an external ID and policy validation | AC-2, AC-6, IA-2 |
| 05 Linux hardening | STIG scan, scripted remediation, and rescan on Ubuntu 22.04 | CM-6, CM-7, AU-2 |
| 06 Azure network security | Segmented VNet, NSG rules, and storage locked to TLS 1.2 | SC-7, SC-8, AC-4 |
Terraform, Rego and OPA Gatekeeper, kind, Trivy, GitHub Actions, AWS CLI, Azure CLI, OpenSCAP.
AWS account IDs and email addresses are replaced with placeholders before anything is committed. Cloud resources are destroyed after each lab; the code rebuilds them.
Built by Ashley Pearce.