Security Digest · Vulnerability record

CVE-2023-37908: org.xwiki.rendering:xwiki-rendering-xml Improper Neutralization of Invalid Characters in Identifiers in Web Pages vulnerability

XWiki Rendering is a generic Rendering system that converts textual input in a given syntax into another syntax.

CVSS 9.1 Critical EPSS 1.1% · p64

Should you act?

Critical, preconditions apply

Severe if exploited, but exploitation needs conditions that may not apply to you. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.

  • Not in CISA KEV, but CISA’s SSVC assessment records a public proof-of-concept, so attack code exists.
  • EPSS 1.1%, percentile 64: FIRST’s model estimates a 1.1% chance of exploitation activity in the next 30 days, higher than 64% of all scored CVEs (moderate). EPSS is a probability, not a verdict.
  • Needs an account (low (any account)). Exposure depends on who can log in.
  • Only applies if your dependency tree includes org.xwiki.rendering:xwiki-rendering-xml (Maven) >= 14.6-rc-1, < 14.10.4, fixed in 14.10.4. Check your lockfile, including transitive dependencies.

What it is

XWiki Rendering is a generic Rendering system that converts textual input in a given syntax into another syntax. The cleaning of attributes during XHTML rendering, introduced in version 14.6-rc-1, allowed the injection of arbitrary HTML code and thus cross-site scripting via invalid attribute names. This can be exploited, e.g., via the link syntax in any content that supports XWiki syntax like comments in XWiki. When a user moves the mouse over a malicious link, the malicious JavaScript code is executed in the context of the user session. When this user is a privileged user who has programming rights, this allows server-side code execution with programming rights, impacting the confidentiality, integrity and availability of the XWiki instance. While this attribute was correctly recognized as not allowed, the attribute was still printed with a prefix data-xwiki-translated-attribute- without further cleaning or validation. This problem has been patched in XWiki 14.10.4 and 15.0 RC1 by removing characters not allowed in data attributes and then validating the cleaned attribute again. There are no known workarounds apart from upgrading to a version including the fix.

From the GitHub advisory

Impact The cleaning of attributes during XHTML rendering, introduced in version 14.6-rc-1, allowed the injection of arbitrary HTML code and thus cross-site scripting via invalid attribute names. This can be exploited, e.g., via the link syntax in any content that supports XWiki syntax like comments in XWiki:

When a user moves the mouse over this link, the malicious JavaScript code is executed in the context of the user session. When this user is a privileged user who has programming rights, this allows server-side code execution with programming rights, impacting the confidentiality, integrity and availability of the XWiki instance.

While this attribute was correctly recognized as not allowed, the attribute was still printed with a prefix data-xwiki-translated-attribute- without further cleaning or validation.

Note that while versions below 14.6 are not vulnerable to this particular vulnerability, they are still vulnerable to XSS through attributes in XWiki syntax, see the corresponding advisory.

Patches This problem has been patched in XWiki 14.10.4 and 15.0 RC1 by removing characters not allowed in data attributes and then validating the cleaned attribute again.

Weakness type: CWE-83, CWE-79, CWE-86.

Product: org.xwiki.rendering:xwiki-rendering-xml (Maven) · Assigned by: GitHub_M · Published: 2023-10-25

Affected and fixed versions

EcosystemPackageVulnerableFirst fixed
Mavenorg.xwiki.rendering:xwiki-rendering-xml>= 14.6-rc-1, < 14.10.414.10.4
VendorProductAffected versions (per CVE record)
xwikixwiki-rendering>= 14.6-rc-1, < 14.10.4

Severity scores

Scored byCVSSScoreVector
GitHub_M3.19.1 CriticalCVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H
GitHub3.19.6 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
  • Attack vector: network (remotely reachable)
  • Attack complexity: low
  • Privileges required: low (any account)
  • User interaction: required (a user must click/open something)
  • Scope: changed (can affect other components)

CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.

Exploitation and urgency

  • CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
  • CISA SSVC assessment: exploitation poc, automatable no, technical impact total.
  • FIRST EPSS (model of 2026-10-21): 1.06% probability of exploitation activity in the next 30 days; percentile 64.

Timeline

  • — CVE record published by GitHub_M
  • — GitHub-reviewed advisory GHSA-663w-2xp3-5739 published (critical)
  • — CVE record last updated

Sources and references

Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.

Vulnerability records

  1. GitHub Advisory GHSA-663w-2xp3-5739 github.com
  2. CVE.org record CVE-2023-37908 cve.org
  3. NVD entry CVE-2023-37908 nvd.nist.gov

Other references from the CVE record and advisory

  1. github.com github.com
  2. github.com github.com
  3. github.com github.com
  4. jira.xwiki.org jira.xwiki.org

Critical Maven advisories · Today’s Security Digest