> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/artifact-registry/3.0/security-and-compliance/vulnerability-scanning.md).

# Vulnerability scanning

View, filter, and act on vulnerability scan results for container images stored in Harness Artifacts, including severity and remediation.

Harness automatically scans every container image you push to a Docker or Helm registry that has security scanning enabled. Results appear on the **Vulnerabilities** tab of the artifact detail page and in the **Vulnerabilities** column on the Artifacts list. This page explains how to read scan results, filter findings by severity, understand what each scanner detects, and decide what to do with a flagged image.

## Before you begin

* Security scanning must be enabled on the registry. Go to [Configure security scanning](https://developer.harness.io/artifact-registry/3.0/registries/configure-security-scanning.md) to enable it.
* Viewing the **Vulnerabilities** tab requires a Harness Security Testing Orchestration (STO) license.
* Scanning applies to Docker and Helm registries only. Other artifact formats do not produce vulnerability findings.

## Find scan results

### On the Artifacts list

The **Artifacts** list includes a **Vulnerabilities** column that shows the finding count for each artifact, broken down by severity: **Critical**, **High**, **Medium**, and **Low**. The **Evaluation Status** column shows one of two states:

| Status          | Meaning                                                                                                                                                         |
| --------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Scanned**     | The scan pipeline ran and produced results. The severity counts reflect the current findings.                                                                   |
| **Not scanned** | The scan pipeline has not run for this artifact, either because scanning was enabled after the artifact was pushed, or the scan pipeline has not yet completed. |

### On the artifact detail page

Click an artifact to open its detail page, then select the **Vulnerabilities** tab. The tab shows every CVE found in the image with the following fields:

| Column                | What it shows                                                                                   |
| --------------------- | ----------------------------------------------------------------------------------------------- |
| **CVE ID**            | The CVE identifier from the National Vulnerability Database (NVD), linked to the full advisory. |
| **Severity**          | Critical, High, Medium, or Low, derived from the CVSS score.                                    |
| **Package**           | The software package or OS component that contains the vulnerability.                           |
| **Installed version** | The version of the package present in the image.                                                |
| **Fixed in**          | The earliest version of the package that resolves the CVE. Blank if no fix is available yet.    |
| **Description**       | A brief summary of the vulnerability.                                                           |

## Filter and sort findings

Use the severity filter buttons at the top of the Vulnerabilities tab (**Critical**, **High**, **Medium**, **Low**) to show findings at that severity level or above. Click the active filter again to clear it.

The table sorts by severity descending by default. Critical findings appear first.

## What do the vulnerability severities mean?

Severity reflects the CVSS base score assigned by the scanner:

| Severity     | CVSS score | What it typically means                                                                                |
| ------------ | ---------- | ------------------------------------------------------------------------------------------------------ |
| **Critical** | 9.0–10.0   | Remote code execution or privilege escalation with no authentication required. Prioritize immediately. |
| **High**     | 7.0–8.9    | Significant impact, often requiring some user interaction or local access. Address before deployment.  |
| **Medium**   | 4.0–6.9    | Meaningful risk in specific conditions. Assess based on the image's deployment context.                |
| **Low**      | 0.1–3.9    | Minimal exploitability. Track and patch on a regular maintenance schedule.                             |

A finding with **no fix available** means the upstream component has a known vulnerability, but no patched release has been published yet. Monitor the CVE advisory for updates. The recommended action is to evaluate whether the vulnerable component is reachable in your deployment context.

## Scanner options: Aqua Trivy and Grype

Harness supports two built-in container scanners. Both are configurable per registry in the **Security** section of the **Configuration** tab:

**Aqua Trivy** scans OS packages (Alpine, Debian, Ubuntu, RHEL, CentOS, and others), application language packages (npm, pip, Cargo, Maven, Go modules, NuGet, and others), and configuration files. It checks findings against the NVD, GitHub Advisory Database, and OS-specific advisory databases.

**Grype** specializes in container image scanning and supports a broad set of ecosystems. It checks findings against the Grype vulnerability database, which aggregates NVD, GitHub Security Advisories, and additional sources. Grype tends to produce fewer false positives on application-layer packages.

You select one scanner per registry. Go to [Configure security scanning](https://developer.harness.io/artifact-registry/3.0/registries/configure-security-scanning.md) to change the scanner for a registry.

{% hint style="info" %}
**The scan runs after push**

The scan pipeline runs asynchronously after an image is pushed. There is a brief window between the push completing and the scan result appearing. During this window, the **Evaluation Status** shows **Not scanned**. Results typically appear within a few minutes of the push.
{% endhint %}

## How do I triage a flagged artifact?

The actions available depend on whether the artifact has been quarantined by a policy.

**If the artifact is not quarantined**, it is pullable. Review the findings and choose one of the following actions:

* Rebuild the image with a patched base image or dependency, then push the new version.
* Accept the risk if the vulnerable component is not reachable or exposed in your deployment.
* Add a policy exception if your team has reviewed the finding and determined it does not apply.

**If the artifact is quarantined**, it cannot be pulled until it is released. Go to [Quarantine](https://developer.harness.io/artifact-registry/3.0/security-and-compliance/quarantine.md) to understand how to review findings and release or delete a quarantined artifact.

**For upstream dependencies** fetched from external registries, Dependency Firewall can block packages before they are cached, using synchronous policy evaluation at the upstream proxy boundary. Go to [Dependency Firewall](https://developer.harness.io/artifact-registry/3.0/concepts/dependency-firewall.md) for that use case.

## Scanner project requirements

The scan pipeline runs inside a Harness project. The project context depends on the registry scope:

| Registry scope         | Scan project                                                                                                                                                                                |
| ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Project-level**      | The scan pipeline runs in the same project as the registry. No additional setup required.                                                                                                   |
| **Organization-level** | Create a project named `default_project` in the organization that contains the registry. The scan pipeline uses this project automatically.                                                 |
| **Account-level**      | Create a project named `default_project` in the `default` organization (created automatically with your account). The scan pipeline uses this project for all account-level registry scans. |

## Troubleshooting

<details>

<summary>The Vulnerabilities tab is not visible on the artifact detail page</summary>

The **Vulnerabilities** tab requires a Harness Security Testing Orchestration (STO) license. If the tab does not appear, your account may not have STO enabled. Contact your Harness account administrator to confirm your license.

</details>

<details>

<summary>The Vulnerabilities tab is visible but shows no findings</summary>

This happens in two situations:

* The scan pipeline ran and found no vulnerabilities. The image is clean according to the configured scanner.
* The scan pipeline has not yet run. Check the **HARNESS ARTIFACT SCAN PIPELINE** in the Harness Pipelines view for the project associated with your registry. If the pipeline has not executed, verify that security scanning is enabled on the registry and that the project context is correctly configured.

</details>

<details>

<summary>The Evaluation Status shows "Not scanned" for an artifact that was pushed a while ago</summary>

Security scanning must be enabled on the registry before images are pushed for results to appear. If scanning was enabled after an image was already in the registry, that image will not be retroactively scanned. Push a new version of the image to trigger a scan, or re-push the existing tag.

</details>

## Next steps

* To enable scanning on a registry, go to [Configure security scanning](https://developer.harness.io/artifact-registry/3.0/registries/configure-security-scanning.md).
* To understand how quarantine works and how to release a quarantined artifact, go to [Quarantine](https://developer.harness.io/artifact-registry/3.0/security-and-compliance/quarantine.md).
* To view the SBOM generated alongside the vulnerability scan, go to [SBOM: view, score, and download](https://developer.harness.io/artifact-registry/3.0/security-and-compliance/sbom-view-score-and-download.md).
* For deeper security orchestration, policy gating, and multi-scanner workflows, go to [Harness Security Testing Orchestration](https://developer.harness.io/security-testing-orchestration/).
* To trigger a rescan on a flagged artifact, go to [Re-evaluate a version](https://developer.harness.io/artifact-registry/3.0/artifacts/re-evaluate-a-version.md).

{% @harness-feedback/feedback %}
