CVE-2026-128138: Red Hat NLTK incomplete fix: additional JVM/JAR sandbox bypass through unvalidated launcher tokens in java command arguments
This Red Hat advisory covers CVE-2026-128138 in Lightspeed Core, OpenShift Lightspeed and Red Hat OpenShift AI (RHOAI).
- CVE-2026-128138
Key facts
- Vendor rating
- High, CVSS 7
- CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H· base score 9.8, scored by GitHub_MAttack vector: network. Attack complexity: low. Privileges required: none. User interaction: none. Scope: unchanged. Confidentiality impact: high. Integrity impact: high. Availability impact: high.- Weakness
- CWE-77: Improper Neutralization of Special Elements used in a Command
- CWE-88: Improper Neutralization of Argument Delimiters in a Command
- CISA SSVC
- Exploitation: proof of concept exists. Automatable: yes. Technical impact: total. CISA, .
- CISA KEV
- Not listed (catalogue checked )
- Fixed in
- 3.10.4
- Source published
- Listed by VulniPulse
Compiled by VulniPulse from the linked sources. How records are collected and checked.
Affected products & platforms
Description
A flaw was found in NLTK. The internal Java execution wrapper fails to properly validate command arguments before launching the Java Virtual Machine (JVM), allowing launcher flags and argument files to be processed.
An attacker who can supply input to these command parameters could exploit this flaw to execute arbitrary code with the permissions of the running application. Red Hat has evaluated this vulnerability as having an Important severity because it could permit arbitrary Java bytecode execution within the execution context of the hosting container or application.
In Red Hat environments, such as Red Hat OpenShift AI and OpenShift Lightspeed, NLTK is integrated as a library supporting natural language processing and machine learning workflows. Exploitation requires an application to pass untrusted input into internal command execution parameters alongside access to attacker-controlled files on the local filesystem.
As a result, the blast radius is constrained to the permissions and lifecycle of the invoking container or application process. Weakness: CWE-88.
Affected Red Hat products: Lightspeed Core; OpenShift Lightspeed; Red Hat OpenShift AI (RHOAI). Red Hat does not currently list a fixing RHSA for this CVE.
Text from the source record.
How to fix CVE-2026-128138
Upgrade to a fixed release named by Red Hat:
- 3.10.4
Vendor guidance
- Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
More Red Hat Linux advisories
CVE-2026-48710May 26, 2026Security restriction bypass via malformed HTTP Host headerCVE-2026-125350Nov 10, 2026Use after free in XMLCVE-2026-114713Nov 8, 2026Command injection via unescaped message tagCVE-2026-90436Nov 6, 2026Remote code execution via insecure deserialization of SOAP fault messagesCVE-2026-123158Nov 5, 2026Password authentication bypass via password hash misinterpretationCVE-2026-115731Nov 3, 2026authenticated remote code execution via report event deserialization
Get an email when Red Hat publishes the next one.
See Red Hat's last 90 days first. Live alerts for one product are free for life; every vendor is US$15 a month after a free month.