Soup Sieve is a CSS selector library for Beautiful Soup. From 0.4 until 3.0, deeply nested CSS pseudo-class selectors can cause recursive parsing to exceed Python’s recursion limit and raise RecursionError . Applications accepting untrusted selectors may experience failure of the affected request or task if the exception is not handled. The demonstrated behavior is recoverable and does not establish an interpreter crash or persistent service-wide denial of service. This vulnerability is fixed in version 3.0.
In Progress Kendo UI for jQuery, the Chart default tooltip renders point values as raw HTML. An attacker who can influence a string point value can inject markup that executes as JavaScript when a user hovers over the point. Successful exploitation may compromise the confidentiality and integrity of data available in the user's browser context. Version 2016.2.607 or earlier may be affected, but earlier releases have not been verified. The latest vulnerable version is 2026.3.1015; the issue is fixed in version 2026.4.1111.
Soup Sieve is a CSS selector library designed to be used with Beautiful Soup 4. Prior to 2.10, matching a developer-authored selector that places the general sibling combinator inside :has() repeatedly scans each candidate element's following siblings when the right-hand side is absent, producing quadratic work on attacker-controlled HTML. Applications that evaluate a fixed selector such as div:has(a ~ b) over untrusted markup can have a worker occupied for minutes by a small document, while the adjacent sibling form and trusted-only document processing are not affected in the same way. This issue is fixed in version 2.10.
In Progress Kendo UI for jQuery, Chart shared-tooltip rendering treats bound category and series-name values as raw HTML. An attacker who can influence values consumed by an affected Chart can inject markup that executes as JavaScript when a user displays the shared tooltip. Successful exploitation may compromise the confidentiality and integrity of data available in the user's browser context. Version 2016.2.607 or earlier may be affected, but earlier releases have not been verified. The latest vulnerable version is 2026.3.1015; the issue is fixed in version 2026.4.1111.
The MongoDB Node.js driver does not consistently enforce its permitted-host restriction during OIDC authentication. A bearer token produced by the embedding application's own login callback may therefore be transmitted to a host outside the configured set of permitted hosts, which may result in unintended disclosure of that credential. No database account is required; the party supplying the connection configuration needs only to influence the settings the application passes to the driver.
In Progress Kendo UI for jQuery, Spreadsheet sheet names are inserted into the SheetsBar menu button's data-sheet-name attribute without encoding. An attacker who can influence a sheet name can inject an event-handler attribute and execute JavaScript when a user hovers over the menu button. Successful exploitation may compromise the confidentiality and integrity of data available in the user's browser context. The first affected version is 2024.4.1112; the latest vulnerable version is 2026.3.1015. The issue is fixed in version 2026.4.1111.
The Plesk GoAccess statistics integration in Plesk Obsidian for Linux, versions 18.0.77 to 18.0.81.7, 18.0.82 to 18.0.82.2, allows cross-subscription information disclosure.
CVE-2026-126837 is a critical vulnerability in the MongoDB Node.js driver affecting versions from 3.3.3 up to but not including 7.8.0. It involves a validation flaw in the DNS SRV host discovery mechanism for mongodb+srv:// connection strings, allowing an attacker who can manipulate DNS SRV responses to redirect connections to arbitrary hosts outside the expected domain. This redirection can lead to unintended disclosure of application data and authentication credentials, as well as altered query results. The vulnerability does not affect mongodb:// connection strings. A fix is available in version 7.8.0.
CVE-2026-126191 is a high-severity vulnerability in the soupsieve CSS selector library used with Beautiful Soup 4. Versions prior to 3.0 contain inefficient regular expression patterns that can cause exponential backtracking when processing certain ambiguous digit escapes in selectors. This can stall worker processes during selector compilation if untrusted selector strings are used. The issue is resolved in soupsieve version 3.0.
CVE-2026-121876 is a vulnerability in Red Hat Enterprise Linux 10 affecting confidential computing guest VM images in OpenShift Sandboxed Containers. The systemd-gpt-auto-generator automatically discovers and mounts GPT partitions at boot without user confirmation. A malicious host can inject swap partitions into the guest's virtual disk, causing the guest kernel to page out TEE-encrypted memory to a host-controlled device, breaking hardware TEE confidentiality guarantees. The vulnerability arises because non-root partitions are still auto-discovered and mounted despite kernel parameters disabling root auto-discovery. Partition table modifications are not detected by remote attestation due to missing PCR 5 measurement. Mitigations include disabling systemd-gpt-auto-generator via kernel command line or masking the generator in the guest image.
CVE-2026-126190 is a high-severity vulnerability in the Soup Sieve CSS selector library used with Beautiful Soup 4. Versions prior to 4.0 suffer from inefficient algorithmic complexity when processing descendant combinators in CSS selectors, causing repeated ancestor walks that increase polynomially with nesting depth. This can lead to significant CPU resource consumption and prolonged worker occupation when processing untrusted HTML or XML documents. The issue is resolved in version 4.0.
CVE-2026-126189 is a vulnerability in the Soup Sieve CSS selector library prior to version 4.0.1. It causes inefficient algorithmic complexity due to repeated subtree searches when evaluating certain selectors, leading to quadratic matching on deeply nested or wide HTML or XML documents. This can result in significant worker occupation time when processing untrusted documents. The issue is fixed in version 4.0.1.
Muse is Meta's AI agent available on iOS, Android, macOS, and web, designed as a personal assistant with extensive access to user data and device functions. Shortly after launch, it was found to access user conversations without explicit permission, likely due to opaque permission controls. A zero-day vulnerability was discovered allowing attackers to redirect Muse's voice recognition server to a malicious endpoint, enabling remote control of the agent with the victim's authorization token. This could allow attackers to access camera, messages, files, and other sensitive data, and perform actions on behalf of the user. Meta has released an update fixing this vulnerability. Additionally, an experimental feature involving human operators handling calls was discontinued after privacy concerns. Users are advised to review and restrict Muse's permissions to mitigate risks.
CVE-2026-124695 is a cross-site scripting (XSS) vulnerability in Progress Software Kendo UI for jQuery. The vulnerability arises because the items[].hint option, documented as a string, is rendered as HTML, allowing attacker-controlled markup to execute as JavaScript when a user views the form. This affects versions from 2020.2.513 through 2026.3.1015. The issue is fixed in version 2026.4.1111.
AppFlowy Cloud versions up to 0.9.64 contain a stored cross-site scripting (XSS) vulnerability. Authenticated workspace members can upload blobs with unvalidated Content-Type headers, which are served inline without Content-Disposition headers or authentication. This allows an attacker to upload HTML or SVG content that is rendered as an active same-origin document via an unauthenticated endpoint, cached for one year, enabling persistent script execution for visitors accessing the URL.
In Progress Kendo UI for jQuery Gantt, the task editor's Parent dropdown renders task titles as HTML. An attacker who can influence a task title shown in the dropdown can cause attacker-controlled markup to execute as JavaScript when a user opens or interacts with the task editor. Successful exploitation may compromise the confidentiality and integrity of data available in the user's browser context. The first affected version is 2020.3.915; the latest vulnerable version is 2026.3.1015. The issue is fixed in version 2026.4.1111.
MLflow versions before 3.18.0 using the FileStore backend contain a path traversal vulnerability. This flaw allows unauthenticated remote attackers to write files outside the intended run directory by exploiting an unvalidated model_id parameter in the log outputs API. Attackers can supply malicious ../ sequences to create directories and write files at arbitrary filesystem locations, constrained only by the server's permissions. FileStore was the default backend before version 3.7.0 and remains opt-in afterward via MLFLOW_ALLOW_FILE_STORE.
CVE-2025-10698 is a vulnerability in OpenText Identity Console versions from 1.8 up to but not including 25.2. It involves incorrect default permissions and a product released in a non-release configuration on Windows, Linux, and 64-bit platforms. This flaw allows unauthorized retrieval of embedded sensitive data. The issue has been fixed in version 25.2.
CVE-2026-16052 is a low-severity injection vulnerability in the Python Software Foundation's CPython implementation. The http.cookies.Morsel class improperly allows semicolons in attribute values and includes them verbatim in output. This flaw could enable an attacker controlling these values to bypass cookie Domain or SameSite restrictions. The issue affects CPython versions prior to 3.16.0 and is fixed in version 3.16.0.
CVE-2026-127030 is a stack-based buffer overflow vulnerability in the EAP-FAST module of FreeRADIUS on Red Hat Enterprise Linux 10. It arises from improper length validation during session ticket processing. A remote, unauthenticated attacker can exploit this by sending a crafted handshake request, causing the FreeRADIUS server process to crash and resulting in a denial of service. EAP-FAST is not enabled by default, so only systems with this module explicitly enabled are affected.