CVE-2026-22728

This report shows a scope-widening issue in the rotate (re-encrypt) flow: the output scope can be derived from untrusted spec.template.metadata.annotations on the input sealed secret.

If a victim sealed secret is strict- or namespace-scoped, an attacker who can submit it to the rotate endpoint can set sealedsecrets.bitnami.com/cluster-wide=true in the template metadata and receive a rotated sealed secret that is cluster-wide, enabling retargeting (metadata.name/metadata.namespace) and unsealing to recover the victim plaintext.

Relevant Links (Pinned)

• Rotate handler uses NewSealedSecret(..., secret) after unsealing: https://github.com/bitnami-labs/sealed-secrets/blob/946bc048f3407117c837da6e4300686522d4c4eb/pkg/controller/controller.go#L560-L606
• Scope derivation reads secret annotations (SecretScope): https://github.com/bitnami-labs/sealed-secrets/blob/946bc048f3407117c837da6e4300686522d4c4eb/pkg/apis/sealedsecrets/v1alpha1/sealedsecret_expansion.go#L112-L122

Root Cause

The rotate flow unseals the input sealed secret to a Secret, then reseals using NewSealedSecret(..., secret).

Because SecretScope(secret) is computed from secret annotations, and unsealing applies spec.template metadata onto the unsealed secret, an attacker can influence the scope of the rotated output by mutating template annotations on the rotate input.

Attack Path

1. Attacker obtains a victim SealedSecret object (for example via read access to resources or logs) and can submit it to the controller rotate endpoint.
2. Attacker sets spec.template.metadata.annotations.sealedsecrets.bitnami.com/cluster-wide=true (and optionally retargets name/namespace fields).
3. Rotate returns a resealed, cluster-wide sealed secret that is no longer bound to the victim name/namespace.
4. Attacker unseals the rotated output in their chosen namespace/name to recover the victim plaintext.

Proof of Concept

Setup + run:

unzip poc.zip -d poc
cd poc
make test

Canonical output (excerpt):

[CALLSITE_HIT]: pkg/apis/sealedsecrets/v1alpha1/sealedsecret_expansion.go:112 SecretScope
[PROOF_MARKER]: scope_widened=true rotated_scope=cluster-wide

Control output (excerpt):

[NC_MARKER]: scope_widened=false strict_scope_preserved=true

Fix Accepted When

Rotate preserves the original sealing scope and does not allow scope widening based on untrusted template metadata; strict or namespace-wide inputs cannot produce cluster-wide outputs.

poc.zip
PR_DESCRIPTION.md
attack_scenario.md

CVE-2026-22728 has a CVSS score of 4.9 (Medium). The vector is network-reachable, high privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. A fixed version is available (0.36.0); upgrading removes the vulnerable code path.

Kodem intelligence

1. What is CVE-2026-22728? CVE-2026-22728 is a medium-severity security vulnerability in github.com/bitnami-labs/sealed-secrets (go), affecting versions < 0.36.0. It is fixed in 0.36.0.
2. How severe is CVE-2026-22728? CVE-2026-22728 has a CVSS score of 4.9 (Medium). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
3. Which versions of github.com/bitnami-labs/sealed-secrets are affected by CVE-2026-22728? github.com/bitnami-labs/sealed-secrets (go) versions < 0.36.0 is affected.
4. Is there a fix for CVE-2026-22728? Yes. CVE-2026-22728 is fixed in 0.36.0. Upgrade to this version or later.
5. Is CVE-2026-22728 exploitable, and should I be worried? Whether CVE-2026-22728 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
6. What actually determines whether CVE-2026-22728 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
7. How do I fix CVE-2026-22728? Upgrade github.com/bitnami-labs/sealed-secrets to 0.36.0 or later.