Summary
Trivy: Helm chart tar bomb causes OOM via unbounded io.ReadAll in parser
When Trivy scans a Helm chart archive (.tgz), its custom tar unpacker reads each entry with io.ReadAll(tr) and no size limit. An attacker who can place a malicious .tgz file in the scanned path can craft a small compressed archive that decompresses to gigabytes, causing the Trivy process to be killed by the OS OOM killer.
Affected configurations
Exploitation requires the attacker to place a crafted .tgz file in a location that Trivy will scan as a Helm chart. This applies to the following scan targets:
| Command | Condition |
|---|---|
trivy config <dir> |
Directory contains a crafted .tgz Helm chart (misconfiguration scanning is always enabled) |
trivy filesystem --scanners misconf <dir> |
Directory contains a crafted .tgz Helm chart and --scanners misconf is explicitly enabled |
trivy image --scanners misconf <image> |
Image contains a crafted .tgz Helm chart and --scanners misconf is explicitly enabled |
Realistic scenarios include:
- A CI pipeline that runs
trivy config .on a repository where a contributor can submit a pull request containing a crafted chart archive. - A pipeline that scans a container image with
--scanners misconf, whose build context includes untrusted.tgzfiles.
Workarounds
If upgrading is not immediately possible:
- Set a memory limit (cgroup/container) on the Trivy process to bound the blast radius.
- Use
--skip-dirsto exclude directories containing untrusted Helm chart archives from the scan. - Avoid scanning repositories or images with untrusted
.tgzfiles.
Credits
Reported by @jamesgol.
Impact
An attacker who satisfies the conditions above can exhaust all available memory on the host running Trivy. The OS OOM killer will terminate the Trivy process and may affect other processes sharing the same host or CI runner.
The practical impact in CI environments is denial of service: the scan fails, the pipeline is blocked, and repeated submissions re-trigger the same condition. Cloud CI runners may also incur additional costs for consumed resources.
There is no impact on confidentiality or integrity of the scanned system.
Crafted input forces the application to consume excessive CPU, memory, or other resources, degrading or denying service. Typical impact: denial of service.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
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. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
Fixed in Trivy v0.71.0 (#10718). The custom tar unpacker was replaced with archive.LoadArchiveFiles from the official helm.sh/helm/v4 SDK, which enforces per-entry and total size limits and validates archive structure. Users should upgrade to v0.71.0 or later.
Frequently Asked Questions
- What is CVE-2026-54448? CVE-2026-54448 is a high-severity uncontrolled resource consumption vulnerability in github.com/aquasecurity/trivy (go), affecting versions < 0.71.0. It is fixed in 0.71.0. Crafted input forces the application to consume excessive CPU, memory, or other resources, degrading or denying service.
- Which versions of github.com/aquasecurity/trivy are affected by CVE-2026-54448? github.com/aquasecurity/trivy (go) versions < 0.71.0 is affected.
- Is there a fix for CVE-2026-54448? Yes. CVE-2026-54448 is fixed in 0.71.0. Upgrade to this version or later.
- Is CVE-2026-54448 exploitable, and should I be worried? Whether CVE-2026-54448 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
- What actually determines whether CVE-2026-54448 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.
- How do I fix CVE-2026-54448? Upgrade
github.com/aquasecurity/trivyto 0.71.0 or later.