The OWASP core rule set (CRS) is a set of generic attack detection rules for use with compatible web application firewalls. Prior to versions 3.3.10, 4.25.1, and 4.28.0, CRS rules in the 921 (protocol-attack), 930 (LFI), 931 (RFI), 932 (RCE), 933 (PHP), 934 (generic), 941 (XSS), 942 (SQLi), and 943 (session-fixation) families inspect XML request bodies via the XPath expression XML:/*. That expression matches only element text nodes — it does not match XML attribute values. As a result, an attacker who places an attack payload inside an XML attribute (e.g. <a href="javascript:alert(1)=window.name"></a> or <a value="' OR 1=1 --"></a>) bypasses CRS detection entirely. This is verified empirically against a local CRS test instance and against the public CRS sandbox. The bypass affects approximately 159 rules across 9 rule files and applies at every paranoia level (PL1 through PL4), since the variable list is more or less shared across PLs. The 944 (Java) family is the only one that correctly includes XML://@* in addition to XML:/* and is therefore unaffected. This issue has been patched in versions 3.3.10, 4.25.1, and 4.28.0.
Severity
CVSS 3.1
7.2 high
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:L/A:N
EPSS
0.4% estimated probability of exploitation activity in the next 30 days; 37th percentile. Scored 2026-11-13. A forecast for this CVE across the internet, not for your own systems.
CWE
CWE-138Improper Neutralization of Special Elements
Show the promptExplain CVE-2026-62971 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-62971?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-62971?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-62971 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.
Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.
Badge for a README or a site
An image that stays current: it is drawn from the data each time it is fetched (a host that proxies images, as GitHub does, may show its copy for a while). Markdown:
From the distributions' data via OSV.dev: the Debian Security Tracker. Modified: reformatted and linked to CVE records. Credits. On the distributions' own pages: Debian.
Attack patterns (CAPEC)
CAPEC-15 Command Delimiters Likelihood: High · Severity: High
CAPEC-34 HTTP Response Splitting Likelihood: Medium · Severity: High
CAPEC-105 HTTP Request Splitting Likelihood: Medium · Severity: High
The attack patterns MITRE's CAPEC relates to this record's weaknesses, as MITRE names them. Credits.