Scanning Kubernetes Manifests With Kubesec

Scanning Kubernetes Manifests with Kubesec

Overview

Kubesec is an open-source security risk analysis tool developed by ControlPlane that inspects Kubernetes resource manifests for common exploitable risks such as privilege escalation, writable host mounts, and excessive capabilities. It assigns a numerical security score to each resource and provides actionable recommendations for hardening. Kubesec can be used as a CLI binary, Docker container, kubectl plugin, admission webhook, or REST API endpoint.

When to Use

• When conducting security assessments that involve scanning kubernetes manifests with kubesec
• When following incident response procedures for related security events
• When performing scheduled security testing or auditing activities
• When validating security controls through hands-on testing

Coverage Gaps & Validation

Kubesec scores a manifest's security context - a positive score is necessary but far from a full assessment:

• It scores what you feed it: Kubesec analyzes the static YAML only. It does not resolve Helm/Kustomize templating, so scan the rendered output (helm template ... | kubesec scan -), not the chart - unrendered {{ }} values evaluate as missing.
• Score is additive, gaps read as silence: a high score from +1 items (readOnlyRootFilesystem, runAsNonRoot, limits) can still sit alongside an unflagged risk; the advise list (ServiceAccountName, AppArmor +3, Seccomp +4) shows controls you don't have. A score: 0 is not a pass.
• Limited resource kinds: Kubesec checks Pod-spec-bearing kinds (Deployment, Pod, DaemonSet, StatefulSet). RBAC, NetworkPolicy, Ingress, CronJob wrappers, and image CVEs are out of scope entirely.
• No image/runtime view: it can't see what's inside the image or what's actually running - only the declared securityContext.

Validate: parse .[0].score and the scoring.advise/critical arrays (-o json | jq), not just the headline message, and gate CI on score < 0 and required advise items being addressed. Scan every document in multi-resource files, run against the rendered manifest, and pair Kubesec with an image scanner (Trivy/Grype) and RBAC review for the risks it structurally cannot see.

Prerequisites

• Kubernetes manifest files (YAML/JSON) for Deployments, Pods, DaemonSets, StatefulSets
• Docker or Go runtime for local installation
• kubectl access for scanning live cluster resources
• CI/CD pipeline access for automated scanning integration

Core Concepts

Security Scoring System

Kubesec assigns a score to each Kubernetes resource based on security checks:

• Positive scores: Awarded for security-enhancing configurations (readOnlyRootFilesystem, runAsNonRoot)
• Zero or negative scores: Indicate missing security controls or dangerous configurations
• Critical advisories: Flagged configurations that represent immediate security risks

Check Categories

1. Privilege Controls: Checks for privileged containers, host PID/network access, root execution
2. Capabilities: Identifies excessive Linux capabilities (SYS_ADMIN, NET_RAW)
3. Volume Mounts: Detects dangerous host path mounts and writable sensitive paths
4. Resource Limits: Validates presence of CPU/memory resource constraints
5. Security Context: Verifies seccomp profiles, AppArmor annotations, SELinux contexts

Installation

Binary Installation

# Linux/macOS
curl -sSL https://github.com/controlplaneio/kubesec/releases/latest/download/kubesec_linux_amd64.tar.gz | \
  tar xz -C /usr/local/bin/ kubesec

# Verify installation
kubesec version

Docker Installation

docker pull kubesec/kubesec:v2

# Scan a manifest file
docker run -i kubesec/kubesec:v2 scan /dev/stdin < deployment.yaml

kubectl Plugin

kubectl krew install kubesec-scan
kubectl kubesec-scan pod mypod -n default

Practical Scanning

Scanning a Single Manifest

# Scan a deployment manifest
kubesec scan deployment.yaml

# Scan with JSON output
kubesec scan -o json deployment.yaml

# Scan from stdin
cat pod.yaml | kubesec scan -

Sample Output

[
  {
    "object": "Pod/web-app.default",
    "valid": true,
    "fileName": "pod.yaml",
    "message": "Passed with a score of 3 points",
    "score": 3,
    "scoring": {
      "passed": [
        {
          "id": "ReadOnlyRootFilesystem",
          "selector": "containers[] .securityContext .readOnlyRootFilesystem == true",
          "reason": "An immutable root filesystem prevents applications from writing to their local disk",
          "points": 1
        },
        {
          "id": "RunAsNonRoot",
          "selector": "containers[] .securityContext .runAsNonRoot == true",
          "reason": "Force the running image to run as a non-root user",
          "points": 1
        },
        {
          "id": "LimitsCPU",
          "selector": "containers[] .resources .limits .cpu",
          "reason": "Enforcing CPU limits prevents DOS via resource exhaustion",
          "points": 1
        }
      ],
      "advise": [
        {
          "id": "ApparmorAny",
          "selector": "metadata .annotations .\"container.apparmor.security.beta.kubernetes.io/nginx\"",
          "reason": "Well defined AppArmor policies reduce the attack surface of the container",
          "points": 3
        },
        {
          "id": "ServiceAccountName",
          "selector": ".spec .serviceAccountName",
          "reason": "Service accounts restrict Kubernetes API access and should be configured",
          "points": 3
        }
      ]
    }
  }
]

Scanning Multiple Resources

# Scan all YAML files in a directory
for file in manifests/*.yaml; do
  echo "=== Scanning $file ==="
  kubesec scan "$file"
done

# Scan multi-document YAML
kubesec scan multi-resource.yaml

Using the HTTP API

# Scan via the public API
curl -sSX POST --data-binary @deployment.yaml \
  https://v2.kubesec.io/scan

# Run a local API server
kubesec http --port 8080 &

# Scan against local server
curl -sSX POST --data-binary @deployment.yaml \
  http://localhost:8080/scan

CI/CD Integration

GitHub Actions

name: Kubesec Scan
on: [pull_request]
jobs:
  kubesec:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - name: Install Kubesec
        run: |
          curl -sSL https://github.com/controlplaneio/kubesec/releases/latest/download/kubesec_linux_amd64.tar.gz | \
            tar xz -C /usr/local/bin/ kubesec
      - name: Scan Manifests
        run: |
          FAIL=0
          for file in k8s/*.yaml; do
            SCORE=$(kubesec scan "$file" | jq '.[0].score')
            echo "$file: score=$SCORE"
            if [ "$SCORE" -lt 0 ]; then
              echo "FAIL: $file has critical issues (score: $SCORE)"
              FAIL=1
            fi
          done
          exit $FAIL

GitLab CI

kubesec-scan:
  stage: security
  image: kubesec/kubesec:v2
  script:
    - |
      for file in k8s/*.yaml; do
        kubesec scan "$file" > /tmp/result.json
        SCORE=$(cat /tmp/result.json | jq '.[0].score')
        if [ "$SCORE" -lt 0 ]; then
          echo "CRITICAL: $file scored $SCORE"
          cat /tmp/result.json | jq '.[0].scoring.critical'
          exit 1
        fi
      done
  artifacts:
    paths:
      - kubesec-results/

Admission Webhook

Deploy Kubesec as a ValidatingWebhookConfiguration to reject insecure manifests at deploy time:

apiVersion: admissionregistration.k8s.io/v1
kind: ValidatingWebhookConfiguration
metadata:
  name: kubesec-webhook
webhooks:
  - name: kubesec.controlplane.io
    rules:
      - apiGroups: [""]
        apiVersions: ["v1"]
        operations: ["CREATE", "UPDATE"]
        resources: ["pods"]
      - apiGroups: ["apps"]
        apiVersions: ["v1"]
        operations: ["CREATE", "UPDATE"]
        resources: ["deployments", "daemonsets", "statefulsets"]
    clientConfig:
      service:
        name: kubesec-webhook
        namespace: kube-system
        path: /scan
    failurePolicy: Fail
    sideEffects: None
    admissionReviewVersions: ["v1"]

Security Checks Reference

Critical Checks (Negative Score)

Best Practice Checks (Positive Score)

References

• Kubesec GitHub Repository
• Kubesec Online Scanner
• ControlPlane Security Tools
• CIS Kubernetes Benchmark
• Kubernetes Pod Security Standards