Skip to content
Kubescape

Kubescape

Category: IaC Security
License: Free (Open-Source, Apache 2.0)
Suphi Cankurt
Suphi Cankurt
AppSec Enthusiast
Updated February 16, 2026
4 min read
Key Takeaways
  • CNCF Incubating project with 11.1k GitHub stars, used by 25,000+ organizations including Intel and AWS for Kubernetes security scanning.
  • Scans manifests, Helm charts, and container images against CIS Benchmarks, NSA-CISA, MITRE ATT&CK, and SOC 2 compliance frameworks using OPA/Rego policies.
  • In-cluster operator provides continuous scanning, eBPF-based runtime threat detection, and automatic network policy generation from observed traffic patterns.
  • Auto-remediation via 'kubescape fix' corrects YAML misconfigurations; 'kubescape vap' generates Validating Admission Policies to block non-compliant deployments.

Kubescape is an open-source Kubernetes security platform. It scans clusters, manifests, and container images for misconfigurations and vulnerabilities. You can run it as a CLI tool or deploy it as an in-cluster operator.

ARMO created Kubescape and donated it to the CNCF, where it sits as an incubating project. 11.1k GitHub stars, 894 forks. More than 25,000 organizations run it, including Intel, AWS, and Bitnami.

What is Kubescape?

It started life as a misconfiguration scanner. Now it validates configurations against CIS Benchmarks, NSA-CISA hardening guidance, and the MITRE ATT&CK framework for containers — plus image scanning, runtime detection, and network policy generation.

CLI scanning
Scans YAML manifests, Helm charts, Kustomize directories, and container images before they reach a cluster. Caches controls locally for offline use.
In-cluster operator
Microservices deployed via Helm that provide continuous scanning, runtime threat detection via eBPF, and network policy recommendations.
Auto-remediation
The fix command corrects misconfigurations in YAML files. The vap command generates Validating Admission Policies to block non-compliant resources at deploy time.

Key features

FeatureDetails
Compliance frameworksCIS Kubernetes Benchmark, NSA-CISA, MITRE ATT&CK, SOC 2
Policy engineOpen Policy Agent (OPA) with Rego controls from public Regolibrary
Image scanningGrype-based CVE detection in base images and dependencies
Image patchingCopacetic rebuilds images with updated packages without changing the application layer
Runtime monitoringeBPF via Inspektor Gadget — watches system calls, network connections, file access
Network policiesObserves real traffic patterns and generates least-privilege NetworkPolicy resources
Risk scoring0–100 per control, with severity threshold flags for CI/CD gating
Output formatsPretty-printed tables, JSON, JUnit XML, SARIF, HTML, PDF
IDE supportVS Code, Lens
CI/CD integrationsGitHub Actions, GitLab CI, Jenkins, CircleCI, pre-commit hooks

Architecture

Kubescape runs in two modes. The CLI is a standalone binary for pre-deployment scanning. The operator is a Helm-deployed set of microservices for continuous in-cluster monitoring.

Kubescape CLI architecture showing scanning workflow from manifests to policy evaluation

Open Policy Agent runs the policy engine. Controls are written in Rego and live in a public repo called Regolibrary. After the first run, the CLI caches controls locally — scans work offline from that point.

Operator components

The operator has four main pieces:

  • kubescape — misconfiguration scanning (same engine as the CLI)
  • kubevuln — image vulnerability scanning using Grype
  • storage — exposes scan data as Kubernetes API objects
  • synchronizer — exports data to external providers

An optional node agent DaemonSet deploys to every node for host scanning and runtime vulnerability relevancy calculations.

Kubescape operator architecture showing in-cluster microservices and data flow

Runtime relevancy
The node agent monitors running workloads via eBPF and feeds data into relevancy calculations. This lets teams focus on vulnerabilities that are actually reachable in production rather than chasing theoretical risks across every image layer.

Image scanning and patching

Image scanning wraps Grype to find CVEs in base images and dependencies. Each scan produces an SBOM and maps vulnerabilities to specific packages.

Copacetic handles automatic patching: it rebuilds images with updated packages without touching the application layer. Run kubescape patch to patch a vulnerable image directly.

Network policy generation

Network policy generation watches real traffic patterns in your cluster and recommends NetworkPolicy resources for least-privilege segmentation. If you’ve ever tried writing network policies by hand, you know why this matters.

Compliance and output

Supported frameworks: CIS Kubernetes Benchmark (multiple versions), NSA-CISA, MITRE ATT&CK for containers, and SOC 2. Custom frameworks can be defined using Rego.

Each control gets a risk score from 0 to 100. The CLI supports namespace filtering with --include-namespaces and --exclude-namespaces, exception rules for risk acceptance, and a severity threshold flag that fails the scan when the score gets too high.

IDE integration

The VS Code extension scans YAML files in the editor and shows findings in the Problems tab. Frameworks download locally for offline scanning. Scans trigger automatically on file save by default.

Kubescape VS Code extension configuration showing framework and scanning options

Offline scanning
Both the CLI and VS Code extension cache compliance frameworks locally after the first download. Useful for air-gapped environments or when you want consistent results across runs without network calls.

Integrations

CI/CD
GitHub Actions GitHub Actions
GitLab CI GitLab CI
Jenkins Jenkins
CircleCI CircleCI
IDE and developer tools
VS Code VS Code
Lens Lens

Getting started

1
Install the CLIcurl -s https://raw.githubusercontent.com/kubescape/kubescape/master/install.sh | /bin/bash or use Homebrew, Krew, Chocolatey, or Scoop.
2
Scan your cluster — Run kubescape scan against your current kubeconfig context. Add framework nsa or framework cis-v1.23 to target a specific standard.
3
Review results — Check the risk score and per-control breakdown. Use --format json --output results.json for machine-readable output, or --format sarif for GitHub code scanning.
4
Deploy the operatorhelm install kubescape kubescape/kubescape-operator -n kubescape --create-namespace for continuous scanning, runtime detection, and network policy generation.

When to use Kubescape

Kubescape fits teams whose security concerns center on Kubernetes. Misconfigurations, image vulnerabilities, runtime threats, compliance — it handles all of these in one open-source tool.

Pick it when you need compliance validation against CIS, NSA-CISA, or MITRE ATT&CK. The in-cluster operator is worth deploying if you want continuous monitoring beyond pre-deployment checks.

For a broader view of Kubernetes security strategy, see our cloud infrastructure security guide. For broader IaC security coverage that includes Terraform and CloudFormation, pair Kubescape with Checkov or Terrascan. If you want a commercial platform with enterprise support, Snyk IaC offers similar Kubernetes scanning inside a larger security suite.

Frequently Asked Questions

What is Kubescape?
Kubescape is an open-source Kubernetes security platform and CNCF incubating project with 11.1k GitHub stars. It scans clusters, manifests, and container images against CIS Benchmarks, NSA-CISA, and MITRE ATT&CK frameworks.
Is Kubescape free to use?
Yes, Kubescape is free and open-source under the Apache 2.0 license. Created by ARMO, it is now a CNCF incubating project trusted by over 25,000 organizations.
What IaC frameworks does Kubescape support?
Kubescape focuses on Kubernetes manifests, Helm charts, Kustomize directories, and container images. It does not scan Terraform or CloudFormation directly. For broader IaC coverage, consider pairing it with Checkov or Terrascan.
Can Kubescape run in CI/CD pipelines?
Yes, Kubescape has official integrations with GitHub Actions, GitLab CI, Jenkins, and CircleCI. It also provides IDE plugins for VS Code and Lens, plus pre-commit hooks for early feedback.
What is the difference between the CLI and operator?
The CLI handles pre-deployment scanning of files and images. The in-cluster operator is a set of microservices deployed via Helm that provide continuous scanning, runtime threat detection via eBPF, and network policy generation inside a running cluster.