A flaw was found in Red Hat OpenShift AI (RHOAI) llama-stack-operator. This vulnerability allows unauthorized access to Llama Stack services deployed in other namespaces via direct network requests, because no NetworkPolicy restricts access to the llama-stack service endpoint. As a result, a user in one namespace can access another user’s Llama Stack instance and potentially view or manipulate sensitive data.

A flaw was found in node-forge. This vulnerability allows unauthenticated attackers to bypass downstream cryptographic verifications and security decisions via crafting ASN.1 (Abstract Syntax Notation One) structures to desynchronize schema validations, yielding a semantic divergence.

A flaw was found in Hugging Face Accelerate. A remote attacker can achieve arbitrary code execution by exploiting a deserialization of untrusted data vulnerability. This occurs when the target visits a malicious page or opens a malicious file, leading to improper validation of user-supplied data during the parsing of checkpoints. Successful exploitation allows the attacker to execute code in the context of the current process.

A flaw was found in runc. This attack is a more sophisticated variant of CVE-2019-16884, which was a flaw that allowed an attacker to trick runc into writing the LSM process labels for a container process into a dummy tmpfs file and thus not apply the correct LSM labels to the container process. The mitigation applied for CVE-2019-16884 was fairly limited and effectively only caused runc to verify that when we write LSM labels that those labels are actual procfs files.

This vulnerability in fontTools varLib allows a crafted .designspace file to trigger arbitrary file writes and XML-based content injection during variable-font generation. Because filenames are not sanitized, an attacker can use path traversal to overwrite files anywhere on the filesystem, and malicious payloads embedded in XML labelname elements can be injected directly into the generated output. When these overwritten files reside in executable or web-served locations, this can enable local remote-code execution or corruption of application or configuration files. The issue affects the varLib CLI and any code that invokes fontTools.varLib.main().

A flaw was found in urllib3 Python library that could lead to a Denial of Service condition. A remote, malicious server can exploit this flaw by responding to a client request with an HTTP message that uses an excessive number of chained compression algorithms. This unlimited decompression chain causes the client system to consume a virtually unbounded amount of CPU resources and memory. The high resource usage leads to service disruption, making the application unresponsive.

A decompression handling flaw has been discovered in urllib3. When streaming a compressed response, urllib3 can perform decoding or decompression based on the HTTP Content-Encoding header (e.g., gzip, deflate, br, or zstd). The library must read compressed data from the network and decompress it until the requested chunk size is met. Any resulting decompressed data that exceeds the requested amount is held in an internal buffer for the next read operation. The decompression logic could cause urllib3 to fully decode a small amount of highly compressed data in a single operation. This can result in excessive resource consumption (high CPU usage and massive memory allocation for the decompressed data; CWE-409) on the client side, even if the application only requested a small chunk of data.

A flaw was found in Fulcio, a free-to-use certificate authority. This vulnerability allows a denial of service (DoS) due to excessive memory allocation when processing a malicious OpenID Connect (OIDC) identity token containing numerous period characters.

A path traversal and arbitrary file overwrite vulnerability has been identified in Argo Workflows during the extraction of archived artifacts, where symbolic links inside a crafted archive are not safely validated before file extraction. An attacker could exploit this flaw by submitting a malicious archive containing symbolic links that point outside the intended extraction directory, causing files to be written or overwritten in unintended locations within the workflow pod. Successful exploitation may allow an attacker to overwrite execution control files and achieve arbitrary command execution during pod startup.

A flaw was found in Expr, an expression language and expression evaluation for Go. This vulnerability allows a denial of service (DoS) via recursive traversal over user-provided deeply nested or cyclic data structures without enforcing a maximum recursion depth, leading to a stack overflow panic and application crash.

A flaw was found in KEDA, a Kubernetes-based Event Driven Autoscaling component. This arbitrary file read vulnerability allows an attacker with permissions to create or modify a TriggerAuthentication resource to read any file from the node's filesystem where the KEDA pod resides. This is due to insufficient path validation when handling Service Account Tokens during HashiCorp Vault authentication. Successful exploitation can lead to the exfiltration of sensitive system information, such as secrets or configuration files.