| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| Duplicate of CVE-2026-50236. This CVE ID was reserved in error for a finding that already had an existing CVE assignment. |
| An out-of-bounds read issue in the Base64 decoder in Amazon aws-sdk-cpp before 1.11.862, on some platforms, might allow a remote authenticated user to crash an application that processes crafted Base64-encoded input.
To remediate this issue, users should upgrade to version 1.11.862. |
| Seroval facilitates JS value stringification, including complex structures beyond JSON.stringify capabilities. Prior to 1.5.3, seroval.fromJSON() allows attacker-controlled JSON Promise control nodes to operate on values from the general deserialization reference table without verifying genuine internal Promise resolver records, causing deserialization side effects with plugins enabled and potentially unintended server-side invocation or remote code execution when downstream frameworks register callable wrappers. This issue is fixed in version 1.5.3. |
| A flaw was found in BlueZ. Insufficient validation of packet length fields in GetFolderItems responses within the Audio/Video Remote Control Profile (AVRCP) implementation allows a malicious Bluetooth device within range to cause an out-of-bounds memory read. This vulnerability, affecting the parse_media_element() and parse_media_folder() functions, can lead to a crash of the bluetoothd daemon, resulting in a Denial of Service (DoS). It could also potentially expose sensitive heap memory contents. Exploitation requires user interaction to pair with the malicious device. |
| Dell Watchdog Timer Driver versions prior to 2.0.0.1 contain an Exposed IOCTL with Insufficient Access Control vulnerability. A low privileged attacker with local access could potentially exploit this vulnerability, leading to Privilege Escalation. |
| ArcadeDB before 26.8.1 fails to bind the authenticated principal (setCurrentUser) on its batch and time-series HTTP handlers. Because no principal is bound on the worker thread, the engine's fine-grained per-type ACL layer (LocalBucket.checkPermissionsOnFile) does not execute for these handlers. In deployments that use per-type or per-group ACLs, a user with database access but only limited per-type permissions can read from and write to types they are not authorized to access by submitting requests to the batch/time-series endpoints. Deployments that rely solely on database-level access control are not affected. |
| Trix is a what-you-see-is-what-you-get rich text editor for everyday writing. Prior to 2.1.17, Trix is vulnerable to cross-site scripting when a data-trix-serialized-attributes attribute bypasses the DOMPurify sanitizer. An attacker can craft HTML containing a data-trix-serialized-attributes attribute with a malicious payload that, when rendered, executes arbitrary JavaScript in the user's session and may perform unauthorized actions or disclose sensitive information. This issue is fixed in version 2.1.17. |
| Valkey is a distributed key-value database. Prior to 7.2.14, 8.0.10, 8.1.9, 9.0.5, and 9.1.1, Valkey's RESTORE command accepts a malformed RDB stream payload that assigns one Pending Entry List NACK to multiple consumers during stream consumer-group deserialization, causing a use-after-free when one consumer is deleted while another still references the shared NACK and potentially allowing remote code execution. This issue is fixed in versions 7.2.14, 8.0.10, 8.1.9, 9.0.5, and 9.1.1. |
| ArcadeDB server (com.arcadedb:arcadedb-server) in versions 26.7.3 and earlier fails to propagate the authenticated principal to asynchronous command worker threads. When an HTTP command is submitted with awaitResponse:false, it executes on an async worker whose DatabaseContext has no bound user, causing the scripting authorization gate to become a no-op. A user with only read access to a single database can submit an asynchronous JavaScript (language=js) command via the /api/v1/command endpoint to run code with unrestricted host access (e.g., database.getSecurity().createUser) and create a server-wide administrator, escalating to full administrative control. Fixed in 26.8.1. |
| The Grav API plugin (getgrav/grav-plugin-api, bundled with Grav 2.0's admin-next/API stack) before version 1.0.14 contains an open redirect weakness in SsoController::sanitizeReturnTo(). The function rejects a literal '//' prefix but does not account for browsers normalizing backslashes to slashes in special (http/https) schemes, so a returnTo value such as '/\evil.com' passes the guard and is later resolved by the browser as the protocol-relative URL '//evil.com'. Following a legitimate OAuth login flow, an attacker-supplied returnTo parameter could redirect an authenticated victim to an attacker-controlled site for post-login phishing. Full browser-side exploitability depends on the admin-next SPA's client-side oauth-callback handler and was not independently verified by the reporter. |
| Grav before 2.0.15 contains a stored cross-site scripting vulnerability in the audio and video media rendering through the sourceParsedownElement method. The media URL fragment is concatenated unescaped into rawHtml source elements, allowing attackers to inject arbitrary HTML and JavaScript that executes in viewers' sessions. |
| Grav before 2.0.15 contains a stored cross-site scripting vulnerability in the detectXss() function where unpaired quotes in unquoted attribute values bypass event-handler detection. Authenticated editors can inject event handlers like onerror= that pass validation and execute in visitor browsers when page content is rendered. |
| SiYuan before v3.7.4 contains an incorrect authorization vulnerability in eight publish-mode reader-facing endpoints that filter results using the visibility list instead of the disabled list. Anonymous visitors can discover and read content from documents explicitly marked as forbidden from publishing by accessing search, backlink, asset content, saved criteria, recent documents, graph, and tag endpoints. |
| Microsoft is aware of an elevation of privilege in the Microsoft Malware Protection Engine in Microsoft Defender publicly referred to as "ShieldBreak ".
We are working to provide a high quality security update that addresses this vulnerability. We will provide information in this CVE when the update is available. |
| MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. Prior to 3.15.0, CreateModelVersion accepts a run_id or model_id after _validate_source_run() or _validate_source_model() in mlflow/server/handlers.py verifies only path containment, allowing authenticated users to create a model version that references another user's artifact directory and read files through GET /model-versions/get-artifact without the required READ permission. This issue is fixed in version 3.15.0. |
| An out-of-bounds read was addressed with improved input validation. This issue is fixed in iOS 26.6.1 and iPadOS 26.6.1, macOS Tahoe 26.6.2. An app may be able to cause unexpected system termination or read kernel memory. |
| The issue was addressed with improved checks. This issue is fixed in iOS 26.6.1 and iPadOS 26.6.1, macOS Tahoe 26.6.2. Processing an image may lead to a denial-of-service. |
| A use after free issue was addressed with improved memory management. This issue is fixed in iOS 26.6.1 and iPadOS 26.6.1, macOS Tahoe 26.6.2. A remote attacker may be able to cause unexpected system termination. |
| The issue was addressed with improved memory handling. This issue is fixed in iOS 18.7.10 and iPadOS 18.7.10, iOS 26.6.1 and iPadOS 26.6.1, macOS Tahoe 26.6.2. Processing maliciously crafted web content may lead to memory corruption. |
| This issue was addressed through improved state management. This issue is fixed in iOS 18.7.10 and iPadOS 18.7.10, iOS 26.6.1 and iPadOS 26.6.1, macOS Tahoe 26.6.2. Processing maliciously crafted web content may lead to an unexpected Safari crash. |
