NLTK (Natural Language Toolkit) is a suite of open source Python modules, data sets, and tutorials supporting research and development in Natural Language Processing. From 3.10.3 until 3.10.4, the NLTK 3.10.3 JVM hardening validates Java launcher tokens in the options channel but appends the caller-controlled cmd sequence without requiring its first token to be a main class. Supplying -jar or an @argfile through nltk.internals.java or StanfordSegmenter.java_class can therefore execute attacker-selected Java bytecode without the unsafe-JAR opt-in. This issue is fixed in version 3.10.4.
Red Hat has issued a critical security update for the java-21-openjdk packages in Red Hat Enterprise Linux 8 and 9. The update addresses multiple vulnerabilities in the OpenJDK 21 Java Runtime Environment and Software Development Kit, including issues related to Kerberos tickets, StringBuilder checks, JFR events, loop handling, vector masks, integer compression, JMX connections, Base64 optimization, and TLS client connections. These fixes resolve potential security weaknesses that could lead to remote code execution and other impacts. Users of affected Red Hat Enterprise Linux versions are advised to apply the update promptly.
The Datadog Agent collects events and metrics from hosts and sends them to Datadog. Prior to datadog-apm-inject 0.66.0, the Datadog Single Step Instrumentation injector stores helper tooling in a predictable host directory. On Linux or Kubernetes systems with SSI enabled and at least one Java service instrumented through SSI, a local unprivileged attacker with shell access to the same host can influence that tooling and execute code with the privileges of the instrumented Java service. Services that use dd-java-agent without SSI and non-Java language runtimes are not affected. This issue is fixed in datadog-apm-inject 0.66.0.
Java4Shell (CVE-2026-114938) is a critical memory safety vulnerability in OpenJDK's Base64 decoding routine on x86-64 CPUs with AVX-512 VBMI support. It allows pre-authentication remote code execution (RCE) in popular enterprise applications such as Jira, Jenkins, and Keycloak by exploiting heap buffer overflows. The flaw can also leak sensitive memory data (JavaBleed), including PII and secret keys. Exploitation complexity is high and requires per-target heap manipulation and ASLR bypass. Oracle and OpenJDK have released patches in November 2026, but some Java distributions may lag in updates. CISA urges immediate patching, especially for internet-exposed systems. No in-the-wild exploitation has been observed yet.
CVE-2026-14265 is a high-severity Java deserialization remote code execution vulnerability in the AWS Advanced JDBC Wrapper, specifically in the RemoteQueryCachePlugin. This plugin, when enabled, deserializes query results from a shared Redis/Valkey cache without class filtering, allowing an attacker with write access to the cache to execute arbitrary code on the application server. The plugin is not enabled by default. A fix is available in version 4.0.1.
CVE-2026-13501 is a command injection vulnerability affecting version 4.0 of the ANTLR4 GoTarget component. The vulnerability is related to improper handling of input that could allow an attacker to execute arbitrary commands. No CVSS score or detailed impact information is provided. No patch or remediation details are available.
LibreOffice allows connecting to databases via JDBC drivers written in Java. A vulnerability exists where a crafted spreadsheet document can specify a driver location from the document itself, causing LibreOffice to load and execute arbitrary Java code when the document is opened. This issue differs from a previously fixed vulnerability (CVE-2026-63277). In fixed versions, the driver location is restricted to user-configured settings, and database connections from pivot tables require explicit user permission to update links.
LibreOffice includes the HSQLDB database engine, which allows SQL statements to invoke Java methods. A vulnerability exists when HSQLDB is accessed via the generic JDBC driver, as it does not restrict which Java classes can be called. This allows a crafted document to execute arbitrary Java code. Fixed versions restrict SQL calls to only built-in HSQLDB functions regardless of access method.
LibreOffice Calc allows linking a cell range to an external data source, with the link saved in the document. A vulnerability exists where opening a crafted document can cause a Java database driver to be loaded from a remote location outside the local machine, enabling execution of arbitrary code. This issue arises because the fix for a previous vulnerability (CVE-2026-63277) did not cover this attack vector. The fix requires that Java class path entries be file URLs pointing only to local files.
A stored cross-site scripting (XSS) vulnerability exists in Qbian61 forum-java between commits 20f7e68 and 63152c4. Authenticated users can inject malicious scripts via unsanitized comment content rendered using jQuery .html() in the reply dialog. Exploitation involves posting comments with img onerror payloads that execute when victims click Reply, allowing theft of non-HttpOnly session cookies and potential account hijacking, including administrator accounts.