The msgpack-python C extension has an unbounded memory leak when deserializing MessagePack data with strict_map_key validation enabled (default). This occurs because the decoded map value is leaked if a map key is rejected, leading to potential denial of service via memory exhaustion. Additionally, the streaming Unpacker API may suffer from a use-after-free due to stale parser-stack references. The issue affects msgpack-python version 1.2.2 and development commits, with fixes available in versions 2.25.0 and 2.29.2_1.
A vulnerability in virtualenv's bash activation prompt allows attacker-controlled prompt text to be interpreted as executable shell syntax under certain older bash compatibility modes and POSIX shells like dash. This occurs when an attacker influences the prompt value passed to virtualenv, and a user subsequently activates the environment and draws the prompt. The issue affects virtualenv versions prior to 2.9.3 and 2.29.3. The vulnerability does not execute code during environment creation but upon prompt display in affected shells. A fix is available in versions 2.9.3 and 2.29.3.
Dell PowerProtect Data Domain, Feature Release versions 7.7.1.0 through 8.8, LTS2024 release versions 7.13.1.0 through 7.13.1.80, LTS2025 release versions 8.3.1.0 through 8.3.1.40, LTS2026 release versions 8.6.1.0 through 8.6.1.20, contain an Improper Validation of Integrity Check Value vulnerability. A high privileged attacker with local access could potentially exploit this vulnerability, leading to Protection mechanism bypass.
The Mang Board plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'align' parameter in all versions up to, and including, 2.4.5 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Because the payload contains no angle brackets, wp_kses_post leaves the shortcode intact on save, allowing the malicious attribute value to survive post submission and fire in any viewer's browser — including an administrator previewing a pending post.
The If-So Dynamic Content – Elementor & All Page Builders Personalization plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'repeater[*][version_uid] (and many sibling fields: freeze-mode, cookie-input, ip-input, compare, utm-value, page-url-compare, group-name, custom, …)' parameter in all versions up to, and including, 1.10.2 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page.
The Postie plugin for WordPress is vulnerable to Stored Cross-Site Scripting via Email Body URL Autolink in all versions up to, and including, 1.9.87 due to insufficient input sanitization and output escaping. This makes it possible for unauthenticated attackers to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. This is possible because the ?postie=get-mail endpoint is reachable by unauthenticated users with no capability check, and the email processing path removes WordPress's kses output filters before saving raw markup, allowing the unescaped href attribute value to be stored as post content.
The s2Member – Excellent for All Kinds of Memberships, Content Restriction Paywalls & Member Access Subscriptions plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'date_format' parameter in all versions up to, and including, <= 261011 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. The payload bypasses wp_kses_post() on save because it contains no literal angle brackets — only backslash-escaped sequences that PHP's date() materializes into HTML tag characters at render time after WordPress's shortcode_parse_atts() applies stripcslashes() to the attribute value.
The WP Visitor Statistics (Real Time Traffic) plugin for WordPress is vulnerable to SQL Injection via the 'from' (and sibling 'to'/'first'/'second') shortcode attribute of the [wsm_showTrafficStatsList] shortcode in versions up to, and including, 8.8 This is due to insufficient escaping on the user-supplied attribute and lack of sufficient preparation on the existing SQL query: `shortcode_atts()` returns attacker-supplied shortcode attributes verbatim (WP core performs no sanitization), only `$atts['id']` is re-sanitized afterward, and the raw value is then concatenated into a double-quoted SQL string literal inside `fnGetHourlyReportByDateNameTimeZone()` (`includes/wsm_db.php:1449`) and executed via `$wpdb->get_results()`. This makes it possible for authenticated attackers, with Contributor-level access and above, to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database.
The GeoDirectory WordPress plugin before 2.8.196 does not properly sanitise or escape a post badge click-through link value before outputting it in an anchor element on a listing, allowing users with the Contributor role and above to perform Stored Cross-Site Scripting attacks against higher-privileged users who view the listing, including an administrator reviewing moderated content.
CVE-2026-104895 is a high-severity vulnerability in the FasterXML jackson-dataformat-smile library's asynchronous Smile parser. It involves unbounded memory allocation based on a length prefix from untrusted input, leading to potential OutOfMemoryError or NegativeArraySizeException crashes. The flaw allows attackers to cause denial of service by submitting crafted Smile data without requiring privileges or user interaction. Fixed versions are available.
CVE-2026-99210 is a vulnerability in the Linux kernel's accel/rocket driver where the rdev pointer is not cleared on device initialization failure. This can cause a later probe to dereference an error pointer, potentially leading to kernel instability or crashes. The issue arises because the rocket_probe() function stores the return value of rocket_device_init() in rdev without checking for errors first. If initialization fails, rdev remains an error pointer, which is then dereferenced improperly.
CVE-2026-99201 is a vulnerability in the Linux kernel's netlink subsystem where the network namespace pointer was hashed in a way that could leak kernel address information. An unprivileged user could exploit this to recover parts of the kernel memory layout. The issue is fixed by replacing the raw pointer with a unique, non-address net_cookie value in the hash key, preventing address leakage. The fix is backported to stable kernels from version 5.15 onward.
CVE-2026-99191 is a vulnerability in the Linux kernel's calipso module affecting the handling of large TCP payload lengths in IPv6 packets. The issue arises from incorrect adjustment of the IPv6 payload length field when processing packets larger than the maximum IPv6 payload length (IPV6_MAXPLEN), causing potential misinterpretation of packet sizes by network subsystems relying on this value. This flaw was fixed by correctly setting the payload length based on the actual socket buffer length, ensuring accurate length representation for large TCP packets with CALIPSO options.
A vulnerability in the Linux kernel's x86 instruction evaluation code caused by a variable shadowing issue in get_seg_reg_override_idx() leads to incorrect handling of segment register overrides. This results in the function always returning a default segment register value, ignoring FS and GS overrides, and causing an infinite loop when a CS override is present. The flaw can cause the kernel to spin indefinitely, preventing user space from regaining control and leaving CPUs tied up until reboot. Additionally, FS or GS prefix usage can cause emulated results to be written to incorrect memory addresses.
A vulnerability in the Linux kernel's SMB client implementation allows improper handling of SetEA requests with extended attribute values that do not fit the request buffer. This flaw occurs because the CIFSSMBSetEA() function copies the extended attribute value into the SMB request buffer without verifying that it fits, potentially causing out-of-bounds memory writes. The issue affects specific Linux kernel versions and has been resolved by adding proper length checks before copying data into the buffer.
WebSub is a W3C recommendation produced by the Social Web Working Group. In recommendation REC-websub-20180123, the intent-verification protocol requires a subscriber to echo the `hub.challenge` value, also exposed as `hub_challenge` by some frameworks, but does not require safe response media typing or disabled MIME sniffing. An attacker can initiate a subscription to obtain a callback URL and redirect a victim browser to a crafted verification request containing HTML or JavaScript. Subscribers that return the challenge as `text/html` can execute the reflected content in the subscriber origin, using a victim's existing session to read private data or perform actions, exposing `localStorage` or `sessionStorage` and browser permissions granted to that origin. An alternative uses the verification response as an attacker-controlled same-origin script resource to bypass `script-src 'self'`; this can remain possible with `application/octet-stream` when MIME sniffing is allowed. Safe media typing and `X-Content-Type-Options: nosniff` are both needed to address the described scenarios. WebSub Rocks demonstrated reflection, but the advisory does not establish practical impact on that demonstration site.
A security update for the Linux kernel in Red Hat Enterprise Linux 6 Extended Lifecycle Support addresses a vulnerability identified as CVE-2026-99999. The issue involves improper validation of the checksum_start value in the IPv4 implementation. Red Hat has released patches to fix this vulnerability and recommends applying the update and rebooting the system to ensure protection. The update is rated as Important by Red Hat Product Security.
MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 3.10.0 until 3.17.0, the prompt branch in _create_model_version() accepts schemeless relative sources because path validation runs only when a source has a URL scheme, and mlflow.prompt.is_prompt is interpreted by tag presence rather than a true value. A remote caller can store a server-local path as the model version artifact location and use /api/2.0/mlflow/model-versions/get-artifact to read files accessible to the MLflow server process. Default unauthenticated servers require no privileges, while basic-auth deployments are exposed to authenticated users who can create registered models. This issue is fixed in version 3.17.0.
follow-redirects is a Node.js HTTP and HTTPS client wrapper that automatically follows redirects. From 1.16.0 until 1.16.2, the documented sensitiveHeaders option is checked with value instanceof Array, so an array created in another JavaScript realm is replaced with an empty list even though Array.isArray recognizes it. When a request then follows an attacker-controlled cross-origin redirect, a custom credential header named in that discarded list can be forwarded to the redirect destination. This issue is fixed in version 1.16.2.
The Model Context Protocol Python SDK is the official Python SDK for Model Context Protocol servers and clients. From 1.10.0 until 1.30.0 and 2.2.0, ClientSession, ClientSessionGroup, and Client tool calls validate server-declared outputSchema documents with jsonschema's default resolver, causing an unresolved $ref value to fetch an HTTP, HTTPS, or file URL from the client's network and filesystem context. A server can trigger blind requests or local-file reads and can stall every session in the process because resolution runs synchronously on the event loop without a timeout. This issue is fixed in versions 1.30.0 and 2.2.0.