TensorFlow v2.18.0 was discovered to output random results when compiling Embedding, leading to unexpected behavior in the application.

pytorch v2.8.0 was discovered to contain an integer overflow in the component torch.nan_to_num-.long().

A syntax error in the component proxy_tensor.py of pytorch v2.7.0 allows attackers to cause a Denial of Service (DoS).

pytorch v2.8.0 was discovered to display unexpected behavior when the components torch.rot90 and torch.randn_like are used together.

An issue in the component torch.linalg.lu of pytorch v2.8.0 allows attackers to cause a Denial of Service (DoS) when performing a slice operation.

PyTorch before 3.7.0 has a bernoulli_p decompose function in decompositions.py even though it lacks full consistency with the eager CPU implementation, negatively affecting nn.Dropout1d, nn.Dropout2d, and nn.Dropout3d for fallback_random=True.

In PyTorch before 2.7.0, bitwise_right_shift produces incorrect output for certain out-of-bounds values of the "other" argument.

In PyTorch before 2.7.0, when torch.compile is used, FractionalMaxPool2d has inconsistent results.

In PyTorch before 2.7.0, when inductor is used, nn.Fold has an assertion error.

In PyTorch through 2.6.0, when eager is used, nn.PairwiseDistance(p=2) produces incorrect results.

apidoc-core is the core parser library to generate apidoc result following the apidoc-spec. A Prototype Pollution vulnerability in the preProcess function of apidoc-core versions thru 0.15.0 allows attackers to inject properties on Object.prototype via supplying a crafted payload, causing denial of service (DoS) as the minimum consequence.

A prototype pollution in the lib.set function of dref v0.1.2 allows attackers to cause a Denial of Service (DoS) via supplying a crafted payload.

Improper Protection Against Voltage and Clock Glitches in FPGA devices, could allow an attacker with physical access to undervolt the platform resulting in a loss of confidentiality.

Malicious code was inserted into the Nx (build system) package and several related plugins. The tampered package was published to the npm software registry, via a supply-chain attack. Affected versions contain code that scans the file system, collects credentials, and posts them to GitHub as a repo under user's accounts.

Flag Forge is a Capture The Flag (CTF) platform. In versions from 2.1.0 to before 2.3.0, the API endpoint GET /api/problems/:id returns challenge hints in plaintext within the question object, regardless of whether the user has unlocked them via point deduction. Users can view all hints for free, undermining the business logic of the platform and reducing the integrity of the challenge system. This issue has been patched in version 2.3.0.

Flag Forge is a Capture The Flag (CTF) platform. In version 2.1.0, the /api/admin/assign-badge endpoint lacks proper access control, allowing any authenticated user to assign high-privilege badges (e.g., Staff) to themselves. This could lead to privilege escalation and impersonation of administrative roles. This issue has been patched in version 2.2.0.

Omni manages Kubernetes on bare metal, virtual machines, or in a cloud. Prior to version 0.48.0, Omni Wireguard SideroLink has the potential to escape. Omni and each Talos machine establish a peer-to-peer (P2P) SideroLink connection using WireGuard to mutually authenticate and authorize access. The WireGuard interface on Omni is configured to ensure that the source IP address of an incoming packet matches the IPv6 address assigned to the Talos peer. However, it performs no validation on the packet's destination address. The Talos end of the SideroLink connection cannot be considered a trusted environment. Workloads running on Kubernetes, especially those configured with host networking, could gain direct access to this link. Therefore, a malicious workload could theoretically send arbitrary packets over the SideroLink interface. This issue has been patched in version 0.48.0.

Claude Code is an agentic coding tool. Prior to Claude Code version 1.0.39, when using Claude Code with Yarn versions 2.0+, Yarn plugins are auto-executed when running yarn --version. This could lead to a bypass of the directory trust dialog in Claude Code, as plugins would be executed prior to the user accepting the risks of working in an untrusted directory. Users running Yarn Classic were unaffected by this issue. This issue has been fixed in version 1.0.39. Users on standard Claude Code auto-update will have received this fix automatically. Users performing manual updates are advised to update to the latest version.

Microsoft Edge (Chromium-based) Remote Code Execution Vulnerability

Binding to an unrestricted ip address in GitHub allows an unauthorized attacker to execute code over a network.

Llama Stack prior to version v0.2.20 accepted unverified parameters in the resolve_ast_by_type function which could potentially allow for remote code execution.

Horilla is a free and open source Human Resource Management System (HRMS). Prior to version 1.4.0, improper sanitization across the application allows XSS via uploaded SVG (and via allowed <embed>), which can be chained to execute JavaScript whenever users view impacted content (e.g., announcements). This can result in admin account takeover. This issue has been patched in version 1.4.0.

Horilla is a free and open source Human Resource Management System (HRMS). Prior to version 1.4.0, the file upload flow performs validation only in the browser and does not enforce server-side checks. An attacker can bypass the client-side validation (for example, with an intercepting proxy or by submitting a crafted request) to store an executable HTML document on the server. When an administrator or other privileged user views the uploaded file, the embedded script runs in their context and sends session cookies (or other credentials) to an attacker-controlled endpoint. The attacker then reuses those credentials to impersonate the admin. This issue has been patched in version 1.4.0.

Improper Input Validation vulnerability in TOTOLINK X6000R allows Command Injection, File Manipulation.This issue affects X6000R: through V9.4.0cu.1360_B20241207.

A vulnerability in Cisco IOS XE Software could allow an authenticated, local attacker with level-15 privileges or an unauthenticated attacker with physical access to an affected device to execute persistent code at boot time and break the chain of trust. This vulnerability is due to improper validation of software packages. An attacker could exploit this vulnerability by placing a crafted file into a specific location on an affected device. A successful exploit could allow the attacker to execute persistent code on the underlying operating system. Because this vulnerability allows an attacker to bypass a major security feature of a device, Cisco has raised the Security Impact Rating (SIR) of this advisory from Medium to High.

Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') vulnerability in TOTOLINK X6000R allows OS Command Injection.This issue affects X6000R: through V9.4.0cu.1360_B20241207.

tar-fs provides filesystem bindings for tar-stream. Versions prior to 3.1.1, 2.1.3, and 1.16.5 are vulnerable to symlink validation bypass if the destination directory is predictable with a specific tarball. This issue has been patched in version 3.1.1, 2.1.4, and 1.16.6. A workaround involves using the ignore option on non files/directories.

Horilla is a free and open source Human Resource Management System (HRMS). A stored cross-site scripting (XSS) vulnerability in Horilla HRM 1.3.0 allows authenticated admin or privileged users to inject malicious JavaScript payloads into multiple fields in the Project and Task modules. These payloads persist in the database and are executed when viewed by an admin or other privileged users through the web interface. Although the issue is not exploitable by unauthenticated users, it still poses a high risk of session hijacking and unauthorized action within high-privilege accounts. At time of publication there is no known patch.

Horilla is a free and open source Human Resource Management System (HRMS). Unauthenticated users can access uploaded resume files in Horilla 1.3.0 by directly guessing or predicting file URLs. These files are stored in a publicly accessible directory, allowing attackers to retrieve sensitive candidate information without authentication. At time of publication there is no known patch.

A vulnerability in the access control list (ACL) programming of Cisco IOS XE Software for Cisco Catalyst 9500X and 9600X Series Switches could allow an unauthenticated, remote attacker to bypass a configured ACL on an affected device. This vulnerability is due to the flooding of traffic from an unlearned MAC address on a switch virtual interface (SVI) that has an egress ACL applied. An attacker could exploit this vulnerability by causing the VLAN to flush its MAC address table. This condition can also occur if the MAC address table is full. A successful exploit could allow the attacker to bypass an egress ACL on an affected device.

A vulnerability in the Day One setup process of Cisco IOS XE Software for Catalyst 9800 Series Wireless Controllers for Cloud (9800-CL) could allow an unauthenticated, remote attacker to access the public-key infrastructure (PKI) server that is running on an affected device. This vulnerability is due to incomplete cleanup upon completion of the Day One setup process. An attacker could exploit this vulnerability by sending Simple Certificate Enrollment Protocol (SCEP) requests to an affected device. A successful exploit could allow the attacker to request a certificate from the virtual wireless controller and then use the acquired certificate to join an attacker-controlled device to the virtual wireless controller.

A vulnerability in the CLI of Cisco IOS XE Software could allow an authenticated, local attacker with administrative privileges to execute arbitrary commands as root on the underlying operating system of an affected device. This vulnerability is due to insufficient validation of user arguments that are passed to specific CLI commands. An attacker could exploit this vulnerability by logging in to the device CLI with valid administrative (level 15) credentials and using crafted commands at the CLI prompt. A successful exploit could allow the attacker to execute arbitrary commands as root.

A vulnerability in the Web Authentication feature of Cisco IOS XE Software could allow an unauthenticated, remote attacker to conduct a reflected cross-site scripting attack (XSS) on an affected device. This vulnerability is due to improper sanitization of user-supplied input. An attacker could exploit this vulnerability by persuading a user to click a malicious link. A successful exploit could allow the attacker to execute a reflected XSS attack and steal user cookies from the affected device.

A vulnerability in the CLI of Cisco IOS Software and Cisco IOS XE Software could allow an authenticated, local attacker to cause an affected device to reload unexpectedly, resulting in a denial of service (DoS) condition. This vulnerability is due to a buffer overflow. An attacker with a low-privileged account could exploit this vulnerability by using crafted commands at the CLI prompt. A successful exploit could allow the attacker to cause the affected device to reload, resulting in a DoS condition.

Multiple vulnerabilities in Cisco IOS XE Software of could allow an authenticated, local attacker with level-15 privileges or an unauthenticated attacker with physical access to the device to execute persistent code at boot time and break the chain of trust. These vulnerabilities are due path traversal and improper image integrity validation. A successful exploit could allow the attacker to execute persistent code on the underlying operating system. Because this allows the attacker to bypass a major security feature of the device, Cisco has raised the Security Impact Rating (SIR) of this advisory from Medium to High. For more information about these vulnerabilities, see the Details ["#details"] section of this advisory. ERP

A vulnerability in the handling of certain Ethernet frames in Cisco IOS XE Software for Catalyst 9000 Series Switches could allow an unauthenticated, adjacent attacker to cause an egress port to become blocked and drop all outbound traffic. This vulnerability is due to improper handling of crafted Ethernet frames. An attacker could exploit this vulnerability by sending crafted Ethernet frames through an affected switch. A successful exploit could allow the attacker to cause the egress port to which the crafted frame is forwarded to start dropping all frames, resulting in a denial of service (DoS) condition.

A vulnerability in the web UI of Cisco IOS Software could allow an authenticated, remote attacker with low privileges to cause a denial of service (DoS) condition on an affected device. This vulnerability is due to improper input validation. An attacker could exploit this vulnerability by sending a crafted URL in an HTTP request. A successful exploit could allow the attacker to cause the affected device to reload, resulting in a DoS condition.

A vulnerability in the Simple Network Management Protocol (SNMP) subsystem of Cisco IOS XE Software could allow an authenticated, remote attacker to cause a denial of service (DoS) condition on an affected device. This vulnerability is due to improper error handling when parsing a specific SNMP request. An attacker could exploit this vulnerability by sending a specific SNMP request to an affected device. A successful exploit could allow the attacker to cause the device to reload unexpectedly, resulting in a DoS condition. This vulnerability affects SNMP versions 1, 2c, and 3. To exploit this vulnerability through SNMPv2c or earlier, the attacker must know a valid read-write or read-only SNMP community string for the affected system. To exploit this vulnerability through SNMPv3, the attacker must have valid SNMP user credentials for the affected system.

A vulnerability in the Simple Network Management Protocol (SNMP) subsystem of Cisco IOS Software and Cisco IOS XE Software could allow the following: An authenticated, remote attacker with low privileges could cause a denial of service (DoS) condition on an affected device that is running Cisco IOS Software or Cisco IOS XE Software. To cause the DoS, the attacker must have the SNMPv2c or earlier read-only community string or valid SNMPv3 user credentials. An authenticated, remote attacker with high privileges could execute code as the root user on an affected device that is running Cisco IOS XE Software. To execute code as the root user, the attacker must have the SNMPv1 or v2c read-only community string or valid SNMPv3 user credentials and administrative or privilege 15 credentials on the affected device. An attacker could exploit this vulnerability by sending a crafted SNMP packet to an affected device over IPv4 or IPv6 networks. This vulnerability is due to a stack overflow condition in the SNMP subsystem of the affected software. A successful exploit could allow a low-privileged attacker to cause the affected system to reload, resulting in a DoS condition, or allow a high-privileged attacker to execute arbitrary code as the root user and obtain full control of the affected system. Note: This vulnerability affects all versions of SNMP.

A vulnerability in the implementation of the TACACS+ protocol in Cisco IOS Software and Cisco IOS XE Software could allow an unauthenticated, remote attacker to view sensitive data or bypass authentication. This vulnerability exists because the system does not properly check whether the required TACACS+ shared secret is configured. A machine-in-the-middle attacker could exploit this vulnerability by intercepting and reading unencrypted TACACS+ messages or impersonating the TACACS+ server and falsely accepting arbitrary authentication requests. A successful exploit could allow the attacker to view sensitive information in a TACACS+ message or bypass authentication and gain access to the affected device.

A vulnerability in the Network-Based Application Recognition (NBAR) feature of Cisco IOS XE Software could allow an unauthenticated, remote attacker to cause an affected device to reload, causing a denial of service (DoS) condition. This vulnerability is due to improper handling of malformed Control and Provisioning of Wireless Access Points (CAPWAP) packets. An attacker could exploit this vulnerability by sending malformed CAPWAP packets through an affected device. A successful exploit could allow the attacker to cause the device to reload unexpectedly, resulting in a DoS condition.

A vulnerability in the HTTP API subsystem of Cisco IOS XE Software could allow a remote attacker to inject commands that will execute with root privileges into the underlying operating system. This vulnerability is due to insufficient input validation. An attacker with administrative privileges could exploit this vulnerability by authenticating to an affected system and performing an API call with crafted input. Alternatively, an unauthenticated attacker could persuade a legitimate user with administrative privileges who is currently logged in to the system to click a crafted link. A successful exploit could allow the attacker to execute arbitrary commands as the root user.

A vulnerability in the access control list (ACL) processing of IPv4 packets of Cisco SD-WAN vEdge Software could allow an unauthenticated, remote attacker to bypass a configured ACL. This vulnerability is due to the improper enforcement of the implicit deny all at the end of a configured ACL. An attacker could exploit this vulnerability by attempting to send unauthorized traffic to an interface on an affected device. A successful exploit could allow the attacker to bypass an ACL on the affected device.

A vulnerability in the IPv6 Router Advertisement (RA) packet processing of Cisco Access Point Software could allow an unauthenticated, adjacent attacker to modify the IPv6 gateway on an affected device. This vulnerability is due to a logic error in the processing of IPv6 RA packets that are received from wireless clients. An attacker could exploit this vulnerability by associating to a wireless network and sending a series of crafted IPv6 RA packets. A successful exploit could allow the attacker to temporarily change the IPv6 gateway of an affected device. This could also lead to intermittent packet loss for any wireless clients that are associated with the affected device.

A vulnerability in the Device Analytics action frame processing of Cisco Wireless Access Point (AP) Software could allow an unauthenticated, adjacent attacker to inject wireless 802.11 action frames with arbitrary information. This vulnerability is due to insufficient verification checks of incoming 802.11 action frames. An attacker could exploit this vulnerability by sending 802.11 Device Analytics action frames with arbitrary parameters. A successful exploit could allow the attacker to inject Device Analytics action frames with arbitrary information, which could modify the Device Analytics data of valid wireless clients that are connected to the same wireless controller.

A security flaw has been discovered in Mangati NovoSGA up to 2.2.9. The impacted element is an unknown function of the file /admin of the component SVG File Handler. Performing manipulation of the argument logoNavbar/logoLogin results in cross site scripting. Remote exploitation of the attack is possible. The exploit has been released to the public and may be exploited.

Integer overflow in V8 in Google Chrome prior to 140.0.7339.207 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)

Integer overflow in V8 in Google Chrome prior to 140.0.7339.207 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)

Side-channel information leakage in V8 in Google Chrome prior to 140.0.7339.207 allowed a remote attacker to leak cross-origin data via a crafted HTML page. (Chromium security severity: High)

Heap buffer overflow in ANGLE in Google Chrome prior to 140.0.7339.185 allowed a remote attacker to potentially exploit heap corruption via malicious network traffic. (Chromium security severity: High)

Use after free in WebRTC in Google Chrome prior to 140.0.7339.185 allowed a remote attacker to potentially exploit heap corruption via a crafted HTML page. (Chromium security severity: High)