A critical security vulnerability in the Wildfly package involves jackson-databind's handling of class-name type IDs. While it validates the outer class, it improperly accepts attacker-selected classes within generic arguments, even if those classes violate declared types or would normally be rejected. This flaw allows for high-impact exploitation affecting confidentiality, integrity, and availability. Versions prior to 42.0.0-r1 are affected, with a fix available in version 42.0.0-r1.
An out-of-bounds write vulnerability was addressed by improved input validation in multiple Apple operating systems. The flaw could allow a user to cause unexpected application termination or potentially execute arbitrary code. The issue is fixed in iOS 26.7.2 and 27.2, iPadOS 26.7.2 and 27.2, macOS Golden Gate 27.2, macOS Sequoia 15.8.2, macOS Tahoe 26.7.2, and tvOS 27.2.
A use-after-free vulnerability in iOS and iPadOS was addressed by improved memory management. The flaw could cause unexpected app termination when processing a maliciously crafted file. The issue is fixed in iOS 27.2 and iPadOS 27.2.
An out-of-bounds read vulnerability was addressed by improved bounds checking in multiple Apple operating systems including iOS, iPadOS, macOS Golden Gate, tvOS, and visionOS version 27.2. The flaw could allow an application to cause unexpected system termination or read kernel memory.
An out-of-bounds write vulnerability in macOS Golden Gate was fixed by improved bounds checking in version 27.2. This flaw could allow an application to cause unexpected system termination or corrupt kernel memory.
A cross-origin security issue in iOS and iPadOS was addressed by improving the tracking of security origins. The vulnerability could allow processing of maliciously crafted web content to disclose sensitive user information. The issue is fixed in iOS 26.7.2 and iPadOS 26.7.2.
An integer overflow vulnerability was addressed through improved input validation in various Apple operating systems. The flaw could cause unexpected application termination when processing a maliciously crafted USD file. The issue is fixed in iOS 27.2, iPadOS 27.2, macOS Golden Gate 27.2, macOS Sequoia 15.8.2, macOS Tahoe 26.7.2, tvOS 27.2, and visionOS 27.2.
A vulnerability in Apple operating systems including iOS, iPadOS, macOS Golden Gate, visionOS, and watchOS could allow an app to access sensitive user data. The issue has been addressed with improved checks and fixed in version 27.2 of these OSes.
A path handling vulnerability in macOS was addressed by improved validation to prevent potential exploitation. The flaw could allow processing of crafted web content to lead to arbitrary code execution. The issue is fixed in macOS Golden Gate 27.2, macOS Sequoia 15.8.2, and macOS Tahoe 26.7.2.
An out-of-bounds read vulnerability was fixed by improving bounds checking in Apple operating systems including iOS, iPadOS, macOS Golden Gate, tvOS, visionOS, and watchOS version 27.2. The flaw could allow processing of a maliciously crafted font to disclose process memory.
A race condition vulnerability in iOS and iPadOS was addressed by adding additional validation. This flaw could allow an app to access sensitive user data. The issue is fixed in iOS 27.2 and iPadOS 27.2.
An out-of-bounds write vulnerability in macOS was addressed by improved bounds checking. This flaw could allow an application to cause unexpected system termination. The issue is fixed in macOS Sequoia 15.8.2 and macOS Tahoe 26.7.2.
An out-of-bounds write vulnerability in macOS was addressed by improved bounds checking. The flaw could allow kernel memory corruption when connecting to a malicious NFS server. Fixed in macOS Golden Gate 27.2, macOS Sequoia 15.8.2, and macOS Tahoe 26.7.2.
probe-image-size versions before 7.5.0 have an unbounded memory allocation vulnerability in the lib/parse_stream/avif.js component. This flaw allows remote attackers to cause a denial of service by supplying an AVIF stream or URL with attacker-declared ftyp or meta box sizes that are excessively large. The parser buffers all received bytes based on these declared sizes, leading to memory exhaustion.
Qwik City versions before 1.20.3 and @qwik.dev/router versions before 2.0.0-rc.4 have a vulnerability in the formToObj function that allows unauthenticated attackers to overwrite built-in function properties by sending specially crafted form POST requests. This causes the Object.prototype.hasOwnProperty.call method to be broken process-wide until the server is restarted.
audio-separator versions before 0.47.2 have an insecure deserialization vulnerability in the handling of Demucs .th model files. This flaw allows attackers to execute arbitrary code by supplying malicious model files loaded with torch.load using weights_only=False. The vulnerability arises when local users can write to the shared /tmp/audio-separator-models/ directory or when attackers control unverified model download sources, leading to code execution upon running Demucs separation.
devpi-web versions before 5.2.0 have an unauthenticated path traversal vulnerability in the +doc documentation route. This flaw allows attackers to craft requests with specially encoded path segments to read arbitrary server files, including sensitive files such as the keyfs sqlite database containing password hashes.
probe-image-size versions before 7.5.0 have an unbounded memory buffering vulnerability in the parseExif function within lib/parse_stream/webp.js. This flaw allows remote attackers to cause memory exhaustion by supplying a WebP stream or URL with an attacker-declared excessively large EXIF chunk length after the VP8X chunk. The vulnerability can lead to denial of service due to process memory exhaustion.
An Observable Discrepancy vulnerability exists in the CAPTCHA module for Drupal 7, versions 7.x-1.0 through 7.x-1.7. This flaw allows unauthenticated attackers to test login credentials despite CAPTCHA protection by exploiting the fact that form validation proceeds even when CAPTCHA responses are missing or incorrect. This behavior can leak feedback that distinguishes invalid credentials from valid credentials paired with failed CAPTCHA responses.