Install this skill
npx skills add mukul975/Anthropic-Cybersecurity-SkillsFramework mappings
MITRE ATT&CK
Overview
Helm is the Kubernetes package manager. Securing Helm deployments requires validating chart provenance, scanning templates for security misconfigurations, enforcing pod security contexts, managing secrets securely, and controlling RBAC for Helm operations.
When to Use
- When deploying or configuring securing helm chart deployments capabilities in your environment
- When establishing security controls aligned to compliance requirements
- When building or improving security architecture for this domain
- When conducting security assessments that require this implementation
Prerequisites
- Helm 3.12+ installed
- kubectl with cluster access
- GnuPG for chart signing/verification
- kubesec or checkov for template scanning
Chart Provenance and Integrity
Sign a Helm Chart
# Generate GPG key for signing
gpg --full-generate-key
# Package and sign chart
helm package ./mychart --sign --key "helm-signing@example.com" --keyring ~/.gnupg/pubring.gpg
# Verify chart signature
helm verify mychart-0.1.0.tgz --keyring ~/.gnupg/pubring.gpgVerify Chart Before Install
# Verify chart from repository
helm pull myrepo/mychart --verify --keyring /path/to/keyring.gpg
# Check chart provenance file
cat mychart-0.1.0.tgz.provTemplate Security Scanning
Render and Scan Templates
# Render templates without deploying
helm template myrelease ./mychart --values values-prod.yaml > rendered.yaml
# Scan with kubesec
kubesec scan rendered.yaml
# Scan with checkov
checkov -f rendered.yaml --framework kubernetes
# Scan with trivy
trivy config rendered.yaml
# Scan with kube-linter
kube-linter lint rendered.yamlHelm Lint for Misconfigurations
# Lint chart
helm lint ./mychart --values values-prod.yaml --strict
# Lint with debug output
helm lint ./mychart --debugSecurity Context Enforcement in values.yaml
# values.yaml - Security hardened defaults
securityContext:
runAsNonRoot: true
runAsUser: 1000
runAsGroup: 3000
fsGroup: 2000
readOnlyRootFilesystem: true
allowPrivilegeEscalation: false
capabilities:
drop:
- ALL
podSecurityContext:
seccompProfile:
type: RuntimeDefault
resources:
limits:
cpu: 500m
memory: 512Mi
requests:
cpu: 100m
memory: 128Mi
networkPolicy:
enabled: true
serviceAccount:
create: true
automountServiceAccountToken: false
image:
pullPolicy: Always
# Use digest instead of tag for immutability
# tag: "1.0.0"
# digest: "sha256:abc123..."Template with Security Contexts
# templates/deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: {{ include "mychart.fullname" . }}
spec:
template:
spec:
automountServiceAccountToken: {{ .Values.serviceAccount.automountServiceAccountToken }}
securityContext:
{{- toYaml .Values.podSecurityContext | nindent 8 }}
containers:
- name: {{ .Chart.Name }}
securityContext:
{{- toYaml .Values.securityContext | nindent 12 }}
image: "{{ .Values.image.repository }}:{{ .Values.image.tag }}"
resources:
{{- toYaml .Values.resources | nindent 12 }}Secrets Management
Use External Secrets (Not Helm Values)
# templates/external-secret.yaml
apiVersion: external-secrets.io/v1beta1
kind: ExternalSecret
metadata:
name: {{ include "mychart.fullname" . }}-secrets
spec:
refreshInterval: 1h
secretStoreRef:
name: aws-secretsmanager
kind: ClusterSecretStore
target:
name: {{ include "mychart.fullname" . }}-secrets
data:
- secretKey: db-password
remoteRef:
key: production/database
property: passwordhelm-secrets Plugin
# Install helm-secrets plugin
helm plugin install https://github.com/jkroepke/helm-secrets
# Encrypt values file
helm secrets encrypt values-secrets.yaml
# Deploy with encrypted secrets
helm secrets install myrelease ./mychart -f values.yaml -f values-secrets.yaml
# Decrypt for editing
helm secrets edit values-secrets.yamlRBAC for Helm Operations
# helm-deployer-role.yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
name: helm-deployer
namespace: production
rules:
- apiGroups: ["", "apps", "batch", "networking.k8s.io"]
resources: ["deployments", "services", "configmaps", "secrets", "ingresses", "jobs"]
verbs: ["get", "list", "create", "update", "patch", "delete"]
- apiGroups: [""]
resources: ["pods", "pods/log"]
verbs: ["get", "list"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
name: helm-deployer-binding
namespace: production
subjects:
- kind: ServiceAccount
name: helm-deployer
namespace: production
roleRef:
kind: Role
name: helm-deployer
apiGroup: rbac.authorization.k8s.ioCI/CD Helm Security Pipeline
# .github/workflows/helm-security.yaml
name: Helm Chart Security
on:
pull_request:
paths: ['charts/**']
jobs:
lint-and-scan:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Helm lint
run: helm lint ./charts/mychart --strict
- name: Render templates
run: helm template test ./charts/mychart -f charts/mychart/values.yaml > rendered.yaml
- name: Scan with kube-linter
uses: stackrox/kube-linter-action@v1
with:
directory: rendered.yaml
- name: Scan with trivy
uses: aquasecurity/trivy-action@master
with:
scan-type: config
scan-ref: rendered.yaml
- name: Scan with checkov
uses: bridgecrewio/checkov-action@master
with:
file: rendered.yaml
framework: kubernetesBest Practices
- Sign charts with GPG and verify before installation
- Render and scan templates before deploying to catch misconfigurations
- Enforce security contexts in values.yaml defaults
- Never store secrets in Helm values - use external secrets or helm-secrets plugin
- Use image digests instead of tags for immutable references
- Restrict Helm RBAC to least privilege per namespace
- Pin chart versions in requirements - never use
latest - Lint strictly in CI with
--strictflag - Review third-party charts before deploying to production
- Use Helm test hooks to validate deployments post-install
Source materials
References and resources
Everything below is rendered for inspection. Script files are read-only and never run.
References 3
api-reference.md1.9 KB
API Reference: Securing Helm Chart Deployments
Helm Security Commands
| Command | Description |
|---|---|
helm lint ./chart --strict |
Lint chart with strict mode |
helm template release ./chart |
Render templates locally |
helm verify chart.tgz |
Verify chart signature |
helm package ./chart --sign --key <key> |
Package and sign |
helm pull repo/chart --verify |
Pull with verification |
Security Context Fields
| Field | Recommended | Description |
|---|---|---|
runAsNonRoot |
true | Prevent root execution |
readOnlyRootFilesystem |
true | Immutable filesystem |
allowPrivilegeEscalation |
false | Block privilege escalation |
capabilities.drop |
[ALL] | Drop all Linux capabilities |
seccompProfile.type |
RuntimeDefault | Syscall filtering |
Security Checks
| Check | Severity | Risk |
|---|---|---|
| Privileged container | High | Full host access |
| hostNetwork enabled | High | Network namespace escape |
| hostPID enabled | High | Process namespace escape |
| :latest image tag | Medium | Non-reproducible builds |
| Missing resource limits | Medium | Resource exhaustion DoS |
| Missing readOnlyRootFilesystem | Medium | Writable filesystem |
Template Scanning Tools
| Tool | Command |
|---|---|
| kubesec | kubesec scan rendered.yaml |
| checkov | checkov -f rendered.yaml --framework kubernetes |
| trivy | trivy config rendered.yaml |
| kube-linter | kube-linter lint rendered.yaml |
Python Libraries
| Library | Version | Purpose |
|---|---|---|
subprocess |
stdlib | Execute helm/kubesec CLI |
re |
stdlib | Pattern matching in rendered YAML |
yaml |
PyYAML >=6.0 | Parse YAML content |
json |
stdlib | Report generation |
References
- Helm Security: https://helm.sh/docs/topics/provenance/
- Helm Secrets Plugin: https://github.com/jkroepke/helm-secrets
- Kubesec: https://kubesec.io/
standards.md1.3 KB
Standards and References - Securing Helm Chart Deployments
NIST SP 800-190
- Section 4.1: Image vulnerabilities and configuration defects
- Section 5.2: Registry security and chart provenance
- Section 5.4: Secure deployment configuration
CIS Kubernetes Benchmark v1.8
- 5.2.1-5.2.9: Pod Security Standards enforced via chart defaults
- 5.7.3: Apply security context to pods and containers
SLSA (Supply chain Levels for Software Artifacts)
- Level 1: Documented build process (Helm chart CI)
- Level 2: Source version controlled, signed provenance
- Level 3: Hardened build platform, signed artifacts
- Level 4: Two-party review, hermetic builds
Helm Security Resources
| Resource | URL |
|---|---|
| Helm Security Best Practices | https://helm.sh/docs/chart_best_practices/ |
| Helm Provenance and Integrity | https://helm.sh/docs/topics/provenance/ |
| kube-linter | https://github.com/stackrox/kube-linter |
| checkov Kubernetes checks | https://www.checkov.io/5.Policy%20Index/kubernetes.html |
| helm-secrets plugin | https://github.com/jkroepke/helm-secrets |
Compliance Mappings
PCI DSS v4.0
- Req 6.3.1: Security vulnerabilities identified and managed
- Req 6.5.1: Changes controlled by change control processes
SOC 2
- CC8.1: Change management - Controlled deployment processes
workflows.md1.1 KB
Workflow - Securing Helm Chart Deployments
Phase 1: Chart Development Security
- Set secure defaults in values.yaml (non-root, read-only fs, resource limits)
- Add network policy templates
- Use external secrets references
- Lint with
helm lint --strict
Phase 2: CI Pipeline
- Render templates:
helm template test ./chart -f values.yaml > rendered.yaml - Lint:
helm lint ./chart --strict - Scan:
kube-linter lint rendered.yaml - Scan:
checkov -f rendered.yaml --framework kubernetes - Sign chart:
helm package ./chart --sign
Phase 3: Deployment
- Verify chart signature:
helm verify chart.tgz - Deploy with production values:
helm install release ./chart -f values-prod.yaml - Verify deployment:
helm test release
Phase 4: Post-Deployment
- Validate security contexts:
kubectl get pods -o jsonpath='{.items[*].spec.securityContext}' - Check network policies applied
- Verify secrets sourced from external store
Phase 5: Maintenance
- Update chart versions in lockfile
- Rescan after dependency updates
- Rotate signing keys annually
Scripts 2
agent.py6.6 KB
Display-only source. This catalog never executes bundled scripts.
#!/usr/bin/env python3
"""Agent for securing Helm chart deployments.
Validates chart provenance, renders and scans templates for
security misconfigurations, checks security contexts, and
enforces Helm deployment security baselines.
"""
import json
import subprocess
import sys
import re
from pathlib import Path
from datetime import datetime
try:
import yaml
except ImportError:
yaml = None
class HelmSecurityAgent:
"""Audits Helm chart deployments for security issues."""
def __init__(self, chart_path, output_dir="./helm_audit"):
self.chart_path = Path(chart_path)
self.output_dir = Path(output_dir)
self.output_dir.mkdir(parents=True, exist_ok=True)
self.findings = []
def _run(self, cmd, timeout=60):
try:
result = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout)
return result.stdout, result.stderr, result.returncode
except (FileNotFoundError, subprocess.TimeoutExpired) as e:
return "", str(e), -1
def lint_chart(self, values_file=None):
"""Run helm lint with strict mode."""
cmd = ["helm", "lint", str(self.chart_path), "--strict"]
if values_file:
cmd.extend(["--values", values_file])
stdout, stderr, rc = self._run(cmd)
if rc != 0:
self.findings.append({
"severity": "medium",
"type": "Lint Failure",
"detail": stderr.strip() or stdout.strip(),
})
return {"returncode": rc, "output": stdout.strip(), "errors": stderr.strip()}
def render_templates(self, values_file=None, release_name="audit"):
"""Render Helm templates without deploying."""
cmd = ["helm", "template", release_name, str(self.chart_path)]
if values_file:
cmd.extend(["--values", values_file])
stdout, stderr, rc = self._run(cmd)
if rc == 0:
rendered_path = self.output_dir / "rendered.yaml"
rendered_path.write_text(stdout)
return str(rendered_path)
return None
def check_security_contexts(self, rendered_path):
"""Check rendered manifests for security context issues."""
issues = []
content = Path(rendered_path).read_text()
checks = [
(r"privileged:\s*true", "high", "Privileged container detected"),
(r"hostNetwork:\s*true", "high", "Host network access enabled"),
(r"hostPID:\s*true", "high", "Host PID namespace shared"),
(r"allowPrivilegeEscalation:\s*true", "medium", "Privilege escalation allowed"),
(r"runAsUser:\s*0\b", "medium", "Running as root (UID 0)"),
]
positive_checks = [
(r"readOnlyRootFilesystem:\s*true", "readOnlyRootFilesystem"),
(r"runAsNonRoot:\s*true", "runAsNonRoot"),
(r"drop:\s*\n\s*-\s*ALL", "drop ALL capabilities"),
]
for pattern, severity, msg in checks:
if re.search(pattern, content):
issues.append({"severity": severity, "issue": msg})
self.findings.append({"severity": severity, "type": "Security Context", "detail": msg})
for pattern, name in positive_checks:
if not re.search(pattern, content):
issues.append({"severity": "medium", "issue": f"Missing: {name}"})
self.findings.append({"severity": "medium", "type": "Missing Hardening", "detail": f"Missing: {name}"})
if "resources:" not in content:
issues.append({"severity": "medium", "issue": "No resource limits defined"})
self.findings.append({"severity": "medium", "type": "Missing Resources", "detail": "No CPU/memory limits"})
return issues
def verify_chart_signature(self, keyring=None):
"""Verify Helm chart provenance signature."""
tgz = list(self.chart_path.parent.glob(f"{self.chart_path.name}-*.tgz"))
if not tgz:
return {"verified": False, "reason": "No packaged chart found"}
cmd = ["helm", "verify", str(tgz[0])]
if keyring:
cmd.extend(["--keyring", keyring])
stdout, stderr, rc = self._run(cmd)
if rc != 0:
self.findings.append({"severity": "medium", "type": "Unsigned Chart", "detail": "Chart signature not verified"})
return {"verified": rc == 0, "output": stdout.strip() or stderr.strip()}
def check_image_references(self, rendered_path):
"""Check if image references use digests or tags."""
content = Path(rendered_path).read_text()
issues = []
for m in re.finditer(r'image:\s*["\']?([^"\'\s]+)["\']?', content):
image = m.group(1)
if ":latest" in image:
issues.append({"image": image, "issue": "Uses :latest tag"})
self.findings.append({"severity": "medium", "type": "Image Tag", "detail": f"latest tag: {image}"})
elif "@sha256:" not in image and ":" not in image:
issues.append({"image": image, "issue": "No tag or digest specified"})
return issues
def scan_with_kubesec(self, rendered_path):
"""Scan rendered templates with kubesec if available."""
stdout, stderr, rc = self._run(["kubesec", "scan", rendered_path])
if rc < 0:
return {"available": False}
try:
return json.loads(stdout)
except json.JSONDecodeError:
return {"error": "Parse failed"}
def generate_report(self, values_file=None):
lint = self.lint_chart(values_file)
rendered = self.render_templates(values_file)
sec_ctx = []
images = []
if rendered:
sec_ctx = self.check_security_contexts(rendered)
images = self.check_image_references(rendered)
report = {
"report_date": datetime.utcnow().isoformat(),
"chart": str(self.chart_path),
"lint_result": lint,
"security_context_issues": sec_ctx,
"image_issues": images,
"findings": self.findings,
"total_findings": len(self.findings),
}
out = self.output_dir / "helm_security_report.json"
with open(out, "w") as f:
json.dump(report, f, indent=2)
print(json.dumps(report, indent=2))
return report
def main():
if len(sys.argv) < 2:
print("Usage: agent.py <chart_path> [--values values.yaml]")
sys.exit(1)
chart = sys.argv[1]
values = None
if "--values" in sys.argv:
values = sys.argv[sys.argv.index("--values") + 1]
agent = HelmSecurityAgent(chart)
agent.generate_report(values)
if __name__ == "__main__":
main()
process.py6.2 KB
Display-only source. This catalog never executes bundled scripts.
#!/usr/bin/env python3
"""
Helm Chart Security Scanner - Render Helm templates and scan
for security misconfigurations in Kubernetes manifests.
"""
import json
import subprocess
import sys
import argparse
import re
from pathlib import Path
SECURITY_CHECKS = [
{
"id": "HELM-001",
"name": "Container runs as root",
"severity": "HIGH",
"pattern": r"runAsNonRoot:\s*false|runAsUser:\s*0",
"remediation": "Set securityContext.runAsNonRoot: true and runAsUser to non-zero",
},
{
"id": "HELM-002",
"name": "Privileged container",
"severity": "CRITICAL",
"pattern": r"privileged:\s*true",
"remediation": "Set securityContext.privileged: false",
},
{
"id": "HELM-003",
"name": "Allow privilege escalation",
"severity": "HIGH",
"pattern": r"allowPrivilegeEscalation:\s*true",
"remediation": "Set securityContext.allowPrivilegeEscalation: false",
},
{
"id": "HELM-004",
"name": "No resource limits",
"severity": "MEDIUM",
"pattern": r"resources:\s*\{\}|resources:\s*null",
"remediation": "Set resources.limits.cpu and resources.limits.memory",
},
{
"id": "HELM-005",
"name": "Uses latest image tag",
"severity": "MEDIUM",
"pattern": r"image:.*:latest|image:\s*[^:]+\s*$",
"remediation": "Use specific image tag or digest instead of :latest",
},
{
"id": "HELM-006",
"name": "HostPath volume mount",
"severity": "HIGH",
"pattern": r"hostPath:",
"remediation": "Avoid hostPath volumes; use PersistentVolumeClaim instead",
},
{
"id": "HELM-007",
"name": "Host network enabled",
"severity": "HIGH",
"pattern": r"hostNetwork:\s*true",
"remediation": "Set hostNetwork: false",
},
{
"id": "HELM-008",
"name": "Host PID namespace",
"severity": "HIGH",
"pattern": r"hostPID:\s*true",
"remediation": "Set hostPID: false",
},
{
"id": "HELM-009",
"name": "Service account token auto-mounted",
"severity": "MEDIUM",
"pattern": r"automountServiceAccountToken:\s*true",
"remediation": "Set automountServiceAccountToken: false unless needed",
},
{
"id": "HELM-010",
"name": "Writable root filesystem",
"severity": "MEDIUM",
"pattern": r"readOnlyRootFilesystem:\s*false",
"remediation": "Set securityContext.readOnlyRootFilesystem: true",
},
]
def render_chart(chart_path: str, values_file: str = None, release_name: str = "scan") -> str:
"""Render Helm chart templates."""
cmd = ["helm", "template", release_name, chart_path]
if values_file:
cmd.extend(["-f", values_file])
result = subprocess.run(cmd, capture_output=True, text=True)
if result.returncode != 0:
print(f"Helm template error: {result.stderr}", file=sys.stderr)
sys.exit(1)
return result.stdout
def scan_rendered(content: str) -> list:
"""Scan rendered YAML for security issues."""
findings = []
lines = content.split("\n")
for check in SECURITY_CHECKS:
for i, line in enumerate(lines, 1):
if re.search(check["pattern"], line):
findings.append({
"id": check["id"],
"name": check["name"],
"severity": check["severity"],
"line": i,
"content": line.strip(),
"remediation": check["remediation"],
})
return findings
def lint_chart(chart_path: str) -> dict:
"""Run helm lint on chart."""
result = subprocess.run(
["helm", "lint", chart_path, "--strict"],
capture_output=True, text=True,
)
return {
"passed": result.returncode == 0,
"output": result.stdout + result.stderr,
}
def generate_report(findings: list, chart_path: str) -> str:
"""Generate security scan report."""
severity_counts = {"CRITICAL": 0, "HIGH": 0, "MEDIUM": 0, "LOW": 0}
for f in findings:
severity_counts[f["severity"]] = severity_counts.get(f["severity"], 0) + 1
report = f"""# Helm Chart Security Scan Report
**Chart:** `{chart_path}`
**Findings:** {len(findings)}
## Summary
| Severity | Count |
|----------|-------|
| CRITICAL | {severity_counts['CRITICAL']} |
| HIGH | {severity_counts['HIGH']} |
| MEDIUM | {severity_counts['MEDIUM']} |
| LOW | {severity_counts['LOW']} |
## Findings
| ID | Severity | Finding | Line | Remediation |
|----|----------|---------|------|-------------|
"""
for f in sorted(findings, key=lambda x: {"CRITICAL": 0, "HIGH": 1, "MEDIUM": 2, "LOW": 3}[x["severity"]]):
report += f"| {f['id']} | {f['severity']} | {f['name']} | {f['line']} | {f['remediation']} |\n"
return report
def main():
parser = argparse.ArgumentParser(description="Helm Chart Security Scanner")
parser.add_argument("chart", help="Path to Helm chart")
parser.add_argument("--values", "-f", help="Values file path")
parser.add_argument("--report", "-r", help="Output report file")
parser.add_argument("--lint", action="store_true", help="Run helm lint")
parser.add_argument("--fail-on", choices=["critical", "high", "medium"],
default="high", help="Fail threshold")
args = parser.parse_args()
if args.lint:
lint_result = lint_chart(args.chart)
print(lint_result["output"])
if not lint_result["passed"]:
print("Lint FAILED")
sys.exit(1)
rendered = render_chart(args.chart, args.values)
findings = scan_rendered(rendered)
report = generate_report(findings, args.chart)
if args.report:
Path(args.report).write_text(report)
print(f"Report written to {args.report}")
else:
print(report)
threshold = {"critical": ["CRITICAL"], "high": ["CRITICAL", "HIGH"],
"medium": ["CRITICAL", "HIGH", "MEDIUM"]}
blocking = [f for f in findings if f["severity"] in threshold[args.fail_on]]
if blocking:
print(f"\nFAILED: {len(blocking)} findings at or above {args.fail_on} severity")
sys.exit(1)
if __name__ == "__main__":
main()
Assets 1
template.mdtext/markdown · 1.0 KBKeep exploring