/

CVE-2026-63459

Severity:

HIGH RISK

(8.7)

Stored XSS (CWE-79) in Vendure RichTextDescriptionCell

CVE-2026-63459

Deferred

Vulnerability type

Category:

Stored Cross-Site Scripting

CWE:

CWE-79

Impact:

Administrative Session Compromise and Cross-Privilege Actions

Severity:

HIGH RISK

(8.7)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Vendure's RichTextDescriptionCell can execute administrator-controlled markup while attempting to convert a description into plain text. A lower-privilege administrator who stores malicious content may cause script execution when another administrator views an affected list row, potentially compromising the viewer's session and administrative privileges.

Affected context

The issue can occur in Vendure deployments that use affected dashboard list views and render administrator-controlled descriptions through RichTextDescriptionCell. Relevant views may include Products, Collections, Promotions, Payment Methods, and Shipping Methods. Risk depends on whether a user can submit description content and whether another administrator later views the stored value.

Exploit preconditions

  • Attack surface: Vendure administrative dashboard list views
  • Required access: Permission to create or modify content displayed in affected description fields
  • Trigger point: Rendering a stored description when an administrator views an affected table row
  • Why exploitation is possible: The dashboard assigns untrusted markup to a live element's innerHTML; active resource markup may execute during assignment before the resulting textContent is read

Why this happens

The vulnerability results from treating DOM insertion as a safe text-extraction mechanism. Assigning untrusted content to innerHTML causes the browser to parse and potentially activate markup before the application reads textContent, so later text extraction does not prevent script execution. This maps to CWE-79: Improper Neutralization of Input During Web Page Generation, commonly known as cross-site scripting.

Potential impact

  • A viewing administrator's session may be compromised
  • Unauthorized administrative actions could be performed in the viewer's context
  • Cross-privilege actions may be possible where authorization boundaries are insufficient
  • Cross-channel administrative data or configuration could be accessed or modified
  • Malicious content may persist and affect multiple administrators

General mitigation guidance

Mitigation generally focuses on ensuring that untrusted descriptions are handled as data rather than executable markup.

Common defensive patterns include:

  • Avoiding innerHTML for untrusted content, including temporary DOM elements used for text extraction
  • Using text-only DOM APIs or trusted, context-aware sanitization when formatting is required
  • Applying allowlists for permitted markup and attributes
  • Treating administrator-entered content as untrusted regardless of the administrator's role
  • Enforcing least privilege for dashboard users, channels, and administrative operations
  • Applying appropriate browser-side protections such as a restrictive Content Security Policy
  • Preserving output encoding at the final rendering context

How teams detect issues like this

Organizations can identify similar issues through:

  • Static analysis rules for assignments to innerHTML, outerHTML, and related DOM sinks
  • Code review checks for DOM parsing performed on user-controlled or administrator-controlled values
  • Tests that submit markup containing active elements and verify that rendered views remain inert
  • Review of rich-text sanitization policies, including allowed tags, attributes, and URL schemes
  • Runtime monitoring for unexpected administrative actions following dashboard content views
  • CI policy enforcement requiring approved sanitization utilities or safe text APIs

Automated security review tools may help flag dangerous DOM sinks, but findings should be assessed against the actual data flow and rendering context.

Sources

SourceLink
github.comgithub.com/vendurehq/vendure/commit/d7aa42a3f0cb524297a1a2fdf700e4aba9aca684
github.comgithub.com/vendurehq/vendure/releases/tag/v3.6.5
github.comgithub.com/vendurehq/vendure/security/advisories/GHSA-xhq9-whgq-49j5

Frequently asked

questions

Administrators who view affected list rows may be exposed when descriptions contain stored active markup. The practical scope depends on dashboard permissions and which fields are displayed.

No. The content can persist in the application, but an attacker generally needs permission to store or modify a description and must cause another administrator to view the affected content.

Treat description values as untrusted, avoid assigning them to innerHTML, and use text-only APIs or a narrowly scoped, context-aware sanitizer when formatting is necessary.

Establish approved rendering patterns, enforce output encoding and sanitization policies, scan code for dangerous DOM sinks, and limit administrative privileges so stored content cannot cross unnecessary trust boundaries.

Find out what's already
exploitable in your codebase.