Unprompted URL Opening via Remote Development Port Forwarding

In JetBrains IntelliJ IDEA before 2026.3 opening arbitrary URLs without confirmation was possible via the Remote Development port forwarding model

Winter CMS: Reflected XSS through the search query parameter in the backend Table widget

Winter CMS is a content management system built on the Laravel PHP framework. From 1.0.420 until 1.2.14, the backend Table widget partial modules/backend/widgets/table/partials/_table.php emits the search query parameter without HTML encoding inside a script template. A script-closing sequence in a crafted top-level GET URL terminates the raw-text template and causes the remaining input to be parsed as backend markup, even when the widget uses searching: false. Any Table or DataTable page is a sink, including the core Editor Settings route /backend/system/settings/update/winter/backend/editor, and an unauthenticated attacker can induce an authorized backend user to follow the link. Script executing in the victim's backend origin can read the CSRF token and issue credentialed requests with that user's permissions. For the core sink, a victim with backend.manage_editor, normally a Developer-role user or superuser, can edit CMS templates, so the injected script can conditionally chain to server-side code execution; this does not grant permissions the victim lacks and does not apply automatically to every third-party plugin sink. This issue is fixed in version 1.2.14.

Incorrect default permissions on session transcripts and prompt history in Kiro CLI

Incorrect default file permissions in Kiro CLI (formerly Amazon Q Developer CLI) before version 2.27.0 on macOS and Linux might allow other local users on the same host to read a user's session transcripts and prompt history, including any credentials or source code they contain. The CLI created its local session storage files with permissions that made them readable by all local users. To remediate this issue, users should upgrade to Kiro CLI version 2.27.0 or later and launch it once. Users who ran Amazon Q Developer CLI 1.16.0 through 1.19.x should also delete the ~/.aws/amazonq/.cli_bash_history file or restrict it to the owner.

Double free in kvm_vfio_pviommu_attach allows local privilege escalation

In kvm_vfio_pviommu_attach of vfio.c, there is a possible double free due to a logic error in the code. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.

Local Privilege Escalation in Android NFC Service Through Permission Check Bypass

In multiple functions of NfcService.java, there is a possible permission bypass due to a missing permission check. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.

The vulnerability is a Use-After-Free (UAF) in the EVS AIDL HAL, arising from a data race on an unsynchronized `std::weak_ptr` within `EvsEnumerator::openCamera()`. Specifically, concurrent calls from a privileged UID (e.g., `AID_AUTOMOTIVE_EVS`) can cause a torn `weak_ptr` state, leading to a virtual call through a freed object's vtable at `cameraObj->getCameraInfo(&desc)`. This allows for potential arbitrary code execution within the `hal_evs_default` process.

An unprivileged application can exploit a lack of package/signature validation in SystemUI's `RemoteQCController` when resolving a hardcoded `content://` URI. This allows the attacker to squat on a specific authority if the legitimate package is absent, inject spoofed UI elements into the trusted SystemUI Quick Controls panel, and gain a Background Activity Launch (BAL) grant from SystemUI when the user interacts with the spoofed element.

In is_valid_header of timed_cache_item.c, there is a possible out-of-bounds write due to a missing bounds check. This could lead to local escalation of privilege with System execution privileges needed. User interaction is not needed for exploitation.

Code Execution via Terraform Tool Path Setting in IntelliJ IDEA Before 2026.3

In JetBrains IntelliJ IDEA before 2026.3, 2026.2.4 code execution was possible in untrusted projects via the Terraform tool path setting

DOM‑based XSS via OpenAPI Preview Error View in IntelliJ IDEA

In JetBrains IntelliJ IDEA before 2026.3 dOM-based XSS in the OpenAPI preview error view was possible

Code Execution via Script Engine Discovery in Untrusted Projects

In JetBrains IntelliJ IDEA before 2026.3, 2026.2.4, 2026.1.5.1, 2025.3.6.2 code execution was possible in untrusted projects via script engine discovery

IBM MQ is affected by a denial of service vulnerability

IBM MQ 9.2.0.0 through 9.2.0.44 LTS, 9.3.0.0 through 9.3.0.42 LTS, 9.4.0.0 through 9.4.0.26 LTS, 9.4.0.0 through 9.4.5.1 CD, and 10.0.0.0 through 10.0.0.5 LTS and CD could allow a remote authenticated attacker to cause a denial of service and perform limited memory corruption due to an integer overflow.

Execution via Malicious ShellCheck Path in IntelliJ IDEA

In JetBrains IntelliJ IDEA before 2026.3 code execution in untrusted projects via a configured ShellCheck path

Path Traversal via Unpacking Remote Dev Container Archives in IntelliJ IDEA

In JetBrains IntelliJ IDEA before 2026.3 path traversal was possible via unpacking Dev Container archives from a remote host

Insecure Server Certificate Addition During Headless IDE Runs

In JetBrains IntelliJ IDEA before 2026.3 improper checks allowed server certificates to be added to the trust store during headless runs

Unauthenticated External Connections via Run Configurations in IntelliJ IDEA

In JetBrains IntelliJ IDEA before 2026.3, 2026.2.4 unauthenticated connection to an arbitrary host via run configurations was possible

Unauthenticated RMI Endpoint Enables Code Execution in JetBrains IntelliJ IDEA before 2026.3

In JetBrains IntelliJ IDEA before 2026.3 code execution was possible via the unauthenticated JDBC worker RMI endpoint

Forge: Regular expression denial of service in PEM decoding (pem.decode)

Forge (also called node-forge) is a native implementation of Transport Layer Security in JavaScript. Prior to 1.4.1, the affected implementation used pem.decode(), lib/pem.js, missing END line to process the relevant input or security decision. node-forge used polynomial-time regular expressions in pem.decode(), including when searching for a matching END line and parsing encapsulated headers. An unauthenticated attacker able to supply PEM input to any high-level parser could consume excessive CPU with a long malformed message and cause denial of service. This issue is fixed in version 1.4.1.

Forge: verifyCertificateChain() removes trusted CA certificates with the same subject from the CA store

Forge (also called node-forge) is a native implementation of Transport Layer Security in JavaScript. Prior to 1.4.1, the affected implementation used pki.verifyCertificateChain(), pki.createCaStore(), duplicate subject names to process the relevant input or security decision. node-forge modified the CA store while trying certificates that shared a subject name during pki.verifyCertificateChain(). A failed chain verification could permanently remove trusted CA certificates from the store, causing later valid chains to fail and creating a persistent availability impact for applications that reused the store. This issue is fixed in version 1.4.1.

Forge: Denial of Service via infinite loop in BigInteger.modInverse() with negative input (incomplete fix of CVE-2026-33891)

Forge (also called node-forge) is a native implementation of Transport Layer Security in JavaScript. Prior to 1.4.1, the affected implementation used BigInteger.prototype.modInverse(), bnModInverse(), lib/jsbn.js, RSA public exponent e to process the relevant input or security decision. node-forge's bundled jsbn implementation entered an infinite loop when BigInteger.prototype.modInverse() received a negative value or modulus. An attacker who could control the RSA public exponent passed to key generation, or a direct modInverse() operand, could block the Node.js event loop at full CPU. Parsed DER, PEM, JWK, PKCS#12, X.509, and TLS values are unsigned and cannot trigger this issue. This issue is fixed in version 1.4.1.