The odf-kit library versions from 0.6.0 up to but not including 0.14.2 contain an HTML injection vulnerability in the odtToHtml() and odsToHtml() functions. This occurs because style attribute values from ODF documents are inserted into HTML style attributes without proper escaping, allowing an attacker to inject arbitrary HTML and execute scripts (XSS) via crafted .odt or .ods files. The issue affects multiple style attributes and is present in both the ODT and ODS to-HTML renderers. A fix is available in version 0.14.2.
CVE-2026-106554 is an algorithmic complexity vulnerability in the jackson-dataformat-cbor library's CBOR parser. The parser accumulates CBOR semantic tags in a TagList that grows by a fixed increment, causing quadratic-time growth in CPU and garbage collection when processing a long run of tags. This can lead to denial of service by exhausting CPU and GC resources. The issue affects multiple 2.x and 3.x versions of jackson-dataformat-cbor and has been fixed in specific later versions.
A critical vulnerability exists in PyJWT, a library used for JSON Web Token handling, where improper validation of public keys allows an attacker to forge valid authentication tokens. This occurs when applications accept both symmetric and asymmetric signing algorithms, causing the library to misinterpret asymmetric public keys as symmetric secrets. This flaw enables unauthenticated remote attackers to bypass authentication without the private key. Deployments that enforce a single algorithm type or use JSON Web Key sets are not affected. No patch or mitigation meeting Red Hat's criteria is currently available.
CVE-2026-121962 is a heap buffer overflow vulnerability in FreeRDP versions before 3.32.2. It occurs in the transport_default_read_pdu function when processing partial Early User Authorization (EUA) reads during PROTOCOL_HYBRID_EX CredSSP. A malicious RDP server can exploit this by splitting the four-byte EUA result into short reads, causing the stream pointer to advance beyond the allocated 16 KiB buffer and triggering an out-of-bounds write. This can lead to client denial of service.
CVE-2026-118391 is a high-severity vulnerability in elixir-ecto's ecto library versions from 3.4.0 up to but not including 3.14.3. It allows an unauthenticated remote attacker to exhaust memory on both the application and database server by sending requests with distinct JSON keys or list lengths. This triggers unbounded caching of prepared queries and prepared statements, which are never released until process restart.
A critical buffer overflow vulnerability (CWE-787) has been identified in libwebsockets, a lightweight C library for network protocols. This flaw allows writing data beyond the intended buffer boundaries, impacting confidentiality, integrity, and availability. The vulnerability affects versions prior to 4.3.3-10.oe2403sp3. A patch is available in version 4.3.3-10.oe2403sp3. The CVSS 3.1 score is 8.1, indicating high severity.
CVE-2026-103582 is a high-severity vulnerability in Apache ActiveMQ NMS OpenWire Client Library versions up to 2.2.0. It involves unbounded memory allocation due to missing validation of frame sizes and length fields in the OpenWire unmarshalling code. An attacker or compromised broker could send crafted frames causing excessive memory consumption or application crashes, resulting in denial of service. The issue is fixed in version 2.3.0, which adds configurable maximum frame size enforcement and length validation.
The BOSSAI campaign involves a single operator using a mesh of 12 AI agents to conduct persistent attacks across 11 countries. The attacker leveraged both abliterated and commercial AI models in specialized roles such as infrastructure building, vulnerability scanning, and exploitation. Primary targets included emergency services, universities, financial institutions, and government systems in Taiwan, Thailand, and Mexico. Confirmed breaches include exfiltration of medical records from Taipei emergency services, student data from Thai universities, and over 200,000 Mexican residential phone records. Despite attempts to jailbreak commercial AI models, the attacker relied on less effective abliterated models. The investigation uncovered the full operational infrastructure, including command histories and victim data, highlighting both the capabilities and limitations of AI-augmented autonomous exploitation.
The Kali Forms WordPress plugin before version 2.4.29 allows users with author-level access and above to inject arbitrary JavaScript into form button configurations. This JavaScript executes in the browser session of any user who clicks the button, including administrators, potentially leading to cross-site scripting (XSS) attacks.
CVE-2026-87039 is a memory-safety vulnerability in the Zephyr RTOS affecting the I2S driver subsystem. The flaw occurs when the user-mode syscall verifier z_vrfy_i2s_configure() skips critical ownership and sanity checks if the frame_clk_freq field is zero, allowing an unvalidated user pointer to reach kernel driver code. This leads to an arbitrary pointer read/write primitive in kernel context, breaking the user-space isolation boundary. The issue affects Zephyr versions from 4.4.0 up to but not including 4.5.0 and is fixed in version 4.5.0.
A critical out-of-bounds write vulnerability exists in the WebGL component of Google Chrome on Android versions prior to 156.0.8078.47. This flaw allows a remote attacker to execute arbitrary code outside the browser sandbox by convincing a user to visit a crafted HTML page. The vulnerability has a high CVSS score of 9.6, indicating severe impact on confidentiality, integrity, and availability.
A security update for webkit2gtk3 addresses multiple vulnerabilities in the Skia graphics library component used by chromium-browser within WebKitGTK. These include use-after-free, cross-origin data disclosure, information disclosure, integer overflow, out-of-bounds read, memory corruption, and improper input validation issues. The vulnerabilities affect versions of webkit2gtk3 prior to 2.54.2-0.el9_8.2. A fixed version 2.54.2-0.el9_8.2 is available from Red Hat.
The scientific library version 0.3.9.0 introduced an out-of-bounds read vulnerability in its Eq and Ord implementations for numbers with extreme exponents. This occurs due to integer overflow in exponent calculations leading to negative indexing of a cached power-of-ten array, causing a segmentation fault (SIGSEGV). The issue affects comparisons and arithmetic operations on Scientific values with very large or small exponents. Version 0.3.10.0 fixes the problem by using Integer for logarithm calculations and adding bounds checks.
A type confusion vulnerability in the V8 JavaScript engine used by Google Chrome versions prior to 157.0.8103.38 allows a remote attacker to potentially execute arbitrary code within the browser sandbox by crafting a malicious HTML page. This vulnerability has a high severity rating and could lead to full compromise of the affected browser instance.
An integer overflow vulnerability exists in the GPU component of Google Chrome versions prior to 157.0.8103.38. This flaw allows a remote attacker who has already compromised the renderer process to potentially execute arbitrary code outside the browser sandbox by using a specially crafted HTML page. The vulnerability is rated with high severity due to its potential impact on confidentiality, integrity, and availability.
A use-after-free vulnerability in the WebAudio component of Google Chrome prior to version 157.0.8103.38 allows a remote attacker to execute arbitrary code within the browser sandbox via a crafted HTML page. This vulnerability has a high severity rating and can lead to full compromise of confidentiality, integrity, and availability of the affected system.