GHSA-2PC9-4J83-QJMR vLLM affected by RCE via auto_map dynamic module loading during model initialization

Summary vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path to execute at server startup. --- Impact An attacker who can influence the model repo/path local directory or remote...

Arbitrary Code Injection

Overview vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs Affected versions of this package are vulnerable to Arbitrary Code Injection via the automap process during model initialization, even when trustremotecode is false. An attacker can execute arbitrary...

vLLM affected by RCE via auto_map dynamic module loading during model initialization

Summary vLLM loads Hugging Face automap dynamic modules during model resolution without gating on trustremotecode, allowing attacker-controlled Python code in a model repo/path to execute at server startup. --- Impact An attacker who can influence the model repo/path local directory or remote...

VoidLink Linux Malware Framework Built with AI Assistance Reaches 88,000 Lines of Code

The recently discovered sophisticated Linux malware framework known as VoidLink is assessed to have been developed by a single person with assistance from an artificial intelligence AI model. That's according to new findings from Check Point Research, which identified operational security blunder...

Server-side Request Forgery (SSRF)

Overview Affected versions of this package are vulnerable to Server-side Request Forgery SSRF via the createDocWithMd function, where unsanitized input in the markdown parameter is passed to downstream processing functions. An attacker can access arbitrary files on the server or interact with...

Out-of-bounds Read

Overview Affected versions of this package are vulnerable to Out-of-bounds Read via the globalCopyFiles function. An attacker can access sensitive files outside the intended directory by supplying arbitrary file paths to the API endpoint. Remediation Upgrade...

Lightweight LLMs for Network Attack Detection in IoT Networks

The rapid growth of Internet of Things IoT devices has increased the scale and diversity of cyberattacks, exposing limitations in traditional intrusion detection systems. Classical machine learning ML models such as Random Forest and Support Vector Machine perform well on known attacks but requir...

Konica Bizhub Multifunction Printers Insertion of Sensitive Information into Externally-Accessible File or Directory (CVE-2024-51977)

An unauthenticated attacker who can access either the HTTP service TCP port 80, the HTTPS service TCP port 443, or the IPP service TCP port 631, can leak several pieces of sensitive information from a vulnerable device. The URI path /etc/mntinfo.csv can be accessed via a GET request and no...

vLLM code injection vulnerability

vLLM is an open-source LLM-based inference and service engine that features high throughput and efficient memory usage. Versions of vLLM from 0.10.1 to 0.14.0 contained a code injection vulnerability. This vulnerability stemmed from the loading of the Hugging Face automap dynamic module during...

PT-2026-3865

Name of the Vulnerable Software and Affected Versions vLLM versions 0.10.1 through 0.13.x Description vLLM is an inference and serving engine for large language models LLMs. The software loads Hugging Face auto map dynamic modules during model resolution without verifying trust remote code. This...

AZL-74988 CVE-2025-55130 affecting package nodejs for versions less than 20.14.0-13

A flaw in Node.js’s Permissions model allows attackers to bypass --allow-fs-read and --allow-fs-write restrictions using crafted relative symlink paths. By chaining directories and symlinks, a script granted access only to the current directory can escape the allowed path and read sensitive files...

CVE-2025-55132

A flaw in Node.js's permission model allows a file's access and modification timestamps to be changed via futimes even when the process has only read permissions. Unlike utimes, futimes does not apply the expected write-permission checks, which means file metadata can be modified in read-only...

CVE-2025-55132

A flaw in Node.js's permission model allows a file's access and modification timestamps to be changed via futimes even when the process has only read permissions. Unlike utimes, futimes does not apply the expected write-permission checks, which means file metadata can be modified in read-only...

UBUNTU-CVE-2026-21636

A flaw in Node.js's permission model allows Unix Domain Socket UDS connections to bypass network restrictions when --permission is enabled. Even without --allow-net, attacker-controlled inputs such as URLs or socketPath options can connect to arbitrary local sockets via net, tls, or undici/fetch...

CVE-2026-21636

CVE-2026-21636 describes a security flaw in Node.js’s Permissions model where Unix Domain Socket (UDS) connections can bypass network restrictions even when --allow-net is not enabled. Attacker-controlled inputs (e.g., URLs or socketPath) could reach arbitrary local sockets via net, tls, or undic...

CVE-2026-21636

A flaw in Node.js's permission model allows Unix Domain Socket UDS connections to bypass network restrictions when --permission is enabled. Even without --allow-net, attacker-controlled inputs such as URLs or socketPath options can connect to arbitrary local sockets via net, tls, or undici/fetch...

CVE-2026-21636

A flaw in Node.js's permission model allows Unix Domain Socket UDS connections to bypass network restrictions when --permission is enabled. Even without --allow-net, attacker-controlled inputs such as URLs or socketPath options can connect to arbitrary local sockets via net, tls, or undici/fetch...

CVE-2026-21636

A flaw in Node.js's permission model allows Unix Domain Socket UDS connections to bypass network restrictions when --permission is enabled. Even without --allow-net, attacker-controlled inputs such as URLs or socketPath options can connect to arbitrary local sockets via net, tls, or undici/fetch...

CVE-2025-55132

The Connected documents confirm CVE-2025-55132: Node.js’ fs.futimes() can bypass the Read-Only permission model, allowing modification of file timestamps even when a process has only read access. Affected products are Node.js releases in the 20/22/24/25 lineages. Impact is potential log tampering...

AI-supported vulnerability triage with the GitHub Security Lab Taskflow Agent

Triaging security alerts is often very repetitive because false positives are caused by patterns that are obvious to a human auditor but difficult to encode as a formal code pattern. But large language models LLMs excel at matching the fuzzy patterns that traditional tools struggle with, so we at...