EUVD-2026-17975

llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserializetensor skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPHCOMPUTE messages. Combined...

CVE-2026-34159 llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend

llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserializetensor skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPHCOMPUTE messages. Combined...

CVE-2026-34159

The CVE-2026-34159 entry for llama.cpp describes an unauthenticated RCE via the RPC backend: prior to v.b8492, deserialize_tensor() omits bounds validation when tensor.buffer == 0, enabling an attacker to read/write arbitrary process memory through crafted GRAPH_COMPUTE messages. Combined with AL...

llama.cpp 缓冲区错误漏洞

Llama.cpp is a multimodal model developed by Georgi Gerganov. Prior versions of llama.cpp b8492 contained a buffer error vulnerability. This vulnerability stemmed from the deserializetensor function in the RPC backend, which skipped all boundary verifications when the buffer field of the tensor w...

PT-2026-29570

Name of the Vulnerable Software and Affected Versions llama.cpp versions prior to b8492 Description A logic bug in the RPC backend's deserialize tensor function allows an unauthenticated attacker to read and write arbitrary process memory. This occurs because bounds validation is skipped when a...

When Labels Are Scarce: A Systematic Mapping of Label-Efficient Code Vulnerability Detection

Machine-learning-based code vulnerability detection CVD has progressed rapidly, from deep program representations to pretrained code models and LLM-centered pipelines. Yet dependable vulnerability labeling remains expensive, noisy, and uneven across projects, languages, and CWE types, motivating...

forman: Foreman: Remote Code Execution via command injection in WebSocket proxy

A flaw was found in Foreman. A remote attacker could exploit a command injection vulnerability in Foreman's WebSocket proxy implementation. This vulnerability arises from the system's use of unsanitized hostname values from compute resource providers when constructing shell commands. By operating...

forman: Foreman: Remote Code Execution via command injection in WebSocket proxy

A flaw was found in Foreman. A remote attacker could exploit a command injection vulnerability in Foreman's WebSocket proxy implementation. This vulnerability arises from the system's use of unsanitized hostname values from compute resource providers when constructing shell commands. By operating...

CVE-2026-1961 Forman: foreman: remote code execution via command injection in websocket proxy

A flaw was found in Foreman. A remote attacker could exploit a command injection vulnerability in Foreman's WebSocket proxy implementation. This vulnerability arises from the system's use of unsanitized hostname values from compute resource providers when constructing shell commands. By operating...

CVE-2026-1961

A flaw was found in Foreman. A remote attacker could exploit a command injection vulnerability in Foreman's WebSocket proxy implementation. This vulnerability arises from the system's use of unsanitized hostname values from compute resource providers when constructing shell commands. By operating...

CVE-2026-1961 Forman: foreman: remote code execution via command injection in websocket proxy

A flaw was found in Foreman. A remote attacker could exploit a command injection vulnerability in Foreman's WebSocket proxy implementation. This vulnerability arises from the system's use of unsanitized hostname values from compute resource providers when constructing shell commands. By operating...

PT-2026-28315

Name of the Vulnerable Software and Affected Versions Foreman versions prior to 3.16.3 Foreman versions prior to 3.17.2 Foreman versions prior to 3.18.1 Description A flaw exists in Foreman that allows a remote attacker to exploit a command injection vulnerability within the WebSocket proxy...

CVE-2026-23348

A flaw was found in the Linux kernel, specifically within the CXL Compute Express Link and NVDIMM Non-Volatile Dual In-line Memory Module subsystems. A race condition can occur when NVDIMM objects attempt to reprobe after the cxlacpi module is removed, while the nvdimmbus object is missing. This...

CVE-2026-23327

The CVE-2026-23327 issue is a Linux kernel vulnerability in the CXL mailbox driver (cxl/mbox). The root cause is that cxl_payload_from_user_allowed() casts and dereferences the user payload without validating its size, allowing an undersized mailbox command to trigger a read past the allocated bu...

Hardening Confidential Federated Compute against Side-Channel Attacks

In this work, we identify a set of side-channels in our Confidential Federated Compute platform that a hypothetical insider could exploit to circumvent differential privacy DP guarantees. We show how DP can mitigate two of the side-channels, one of which has been implemented in our open-source...

USN-8095-3 linux-realtime, linux-realtime-6.8 vulnerabilities

Qualys discovered that several vulnerabilities existed in the AppArmor Linux kernel Security Module LSM. An unprivileged local attacker could use these issues to load, replace, and remove arbitrary AppArmor profiles causing denial of service, exposure of sensitive information kernel memory, local...

USN-8095-1 linux, linux-aws, linux-gcp, linux-gcp-6.8, linux-gke, linux-gkeop, linux-hwe-6.8, linux-ibm, linux-ibm-6.8, linux-lowlatency, linux-lowlatency-hwe-6.8, linux-oracle, linux-oracle-6.8 vulnerabilities

Qualys discovered that several vulnerabilities existed in the AppArmor Linux kernel Security Module LSM. An unprivileged local attacker could use these issues to load, replace, and remove arbitrary AppArmor profiles causing denial of service, exposure of sensitive information kernel memory, local...

Canada Needs Nationalized, Public AI

Canada has a choice to make about its artificial intelligence future. The Carney administration is investing $2-billion over five years in its Sovereign AI Compute Strategy. Will any value generated by "sovereign AI" be captured in Canada, making a difference in the lives of Canadians, or is this...

VulnCheck KEV: CVE-2025-62593

Ray is an AI compute engine. Prior to version 2.52.0, developers working with Ray as a development tool can be exploited via a critical RCE vulnerability exploitable via Firefox and Safari. This vulnerability is due to an insufficient guard against browser-based attacks, as the current defense us...

Vulnerabilities fixed in Microsoft Azure

Microsoft has fixed vulnerabilities in several Azure components. A malicious party could exploit the vulnerabilities to impersonate another user, grant themselves elevated privileges or gain access to sensitive data. Azure Entra ID: |----------------|------|-------------------------------------| ...