CVE-2023-54274

CVE-2023-54274 affects the Linux kernel RDMA/srpt path. Root cause: during unregistering a MAD agent the code could see a non-null mad_agent that actually held an error value, allowing a NULL dereference in a short race window when srpt_add_one() and srpt_remove_one() run concurrently (e.g., RoCE...

CVE-2023-54274

In the Linux kernel, the following vulnerability has been resolved: RDMA/srpt: Add a check for valid 'madagent' pointer When unregistering MAD agent, srpt module has a non-null check for 'madagent' pointer before invoking ibunregistermadagent. This check can pass if 'madagent' variable holds an...

CVE-2023-54274 RDMA/srpt: Add a check for valid 'mad_agent' pointer

In the Linux kernel, the following vulnerability has been resolved: RDMA/srpt: Add a check for valid 'madagent' pointer When unregistering MAD agent, srpt module has a non-null check for 'madagent' pointer before invoking ibunregistermadagent. This check can pass if 'madagent' variable holds an...

CVE-2023-54274 RDMA/srpt: Add a check for valid 'mad_agent' pointer

In the Linux kernel, the following vulnerability has been resolved: RDMA/srpt: Add a check for valid 'madagent' pointer When unregistering MAD agent, srpt module has a non-null check for 'madagent' pointer before invoking ibunregistermadagent. This check can pass if 'madagent' variable holds an...

Microsoft Azure Guest Agent Installed (Windows)

Binary data microsoftazureguestagentwininstalled.nbin...

Linux kernel 安全漏洞

Linux kernel is the kernel used by Linux, the open source operating system of the Linux Foundation in the United States. A security vulnerability exists in the Linux kernel that stems from a failure to validate the validity of the madagent pointer in RDMA/srpt, which could result in a null pointe...

PT-2025-54103

Name of the Vulnerable Software and Affected Versions Linux kernel affected versions not specified Description The Linux kernel contains a flaw within the RDMA/srpt module related to the handling of MAD Management Agent agent pointers during unregistration. Specifically, a check for a valid 'mad...

CVE-2025-69201

CVE-2025-69201 (Tugtainer) affects the tugtainer-agent API, where the POST api/command/run endpoint is vulnerable to arbitrary argument injection. The issue exists in versions prior to 1.15.1 and is fixed in 1.15.1. Documented impact is remote command execution via the affected endpoint, with the...

CVE-2025-69201 Tugtainer has RCE in Agent Command Execution Api

Tugtainer is a self-hosted app for automating updates of docker containers. In versions prior to 1.15.1, arbitary arguments can be injected in tugtainer-agent POST api/command/run. Version 1.15.1 fixes the issue...

CVE-2025-69201 Tugtainer has RCE in Agent Command Execution Api

Tugtainer is a self-hosted app for automating updates of docker containers. In versions prior to 1.15.1, arbitary arguments can be injected in tugtainer-agent POST api/command/run. Version 1.15.1 fixes the issue...

CVE-2025-69201 Tugtainer has RCE in Agent Command Execution Api

Tugtainer is a self-hosted app for automating updates of docker containers. In versions prior to 1.15.1, arbitary arguments can be injected in tugtainer-agent POST api/command/run. Version 1.15.1 fixes the issue...

EUVD-2025-205598

Tugtainer is a self-hosted app for automating updates of docker containers. In versions prior to 1.15.1, arbitary arguments can be injected in tugtainer-agent POST api/command/run. Version 1.15.1 fixes the issue...

Tugtainer 命令注入漏洞

Tugtainer is an automated Docker container update application with a web UI by the individual developer Eugene Savin. A command injection vulnerability exists in Tugtainer versions prior to 1.15.1, which stems from the POST api/command/run interface of tugtainer-agent that can inject arbitrary...

pangolin 安全漏洞

pangolin is an agent software from the Pangolin open source. A security vulnerability exists in pangolin 1.6.2 and earlier versions, which stems from a problem with the 2FA component that could lead to elevation of privilege...

Toward Trustworthy Agentic AI: A Multimodal Framework for Preventing Prompt Injection Attacks

Powerful autonomous systems, which reason, plan, and converse using and between numerous tools and agents, are made possible by Large Language Models LLMs, Vision-Language Models VLMs, and new agentic AI systems, like LangChain and GraphChain. Nevertheless, this agentic environment increases the...

PT-2025-53753

Name of the Vulnerable Software and Affected Versions Tugtainer versions prior to 1.15.1 Description Tugtainer is a self-hosted application designed for automating updates of docker containers. A flaw exists where arbitrary arguments can be injected. This occurs through the POST api/command/run...

Security update for buildah

This update for buildah fixes the following issues: CVE-2025-47914: golang.org/x/crypto/ssh/agent: Fixed out of bounds read caused by non validated message size bsc1254054 CVE-2025-47913: golang.org/x/crypto/ssh/agent: Fixed client process termination when receiving an unexpected message type in...

SUSE-SU-2025:4526-1 Security update for buildah

This update for buildah fixes the following issues: - CVE-2025-47914: golang.org/x/crypto/ssh/agent: Fixed out of bounds read caused by non validated message size bsc1254054 - CVE-2025-47913: golang.org/x/crypto/ssh/agent: Fixed client process termination when receiving an unexpected message type...

blacksmith

BlacksmithAI OPEN-SOURCE AI-Powered Penetration Testing F...

Analyzing Code Injection Attacks on LLM-Based Multi-Agent Systems in Software Development

Agentic AI and Multi-Agent Systems are poised to dominate industry and society imminently. Powered by goal-driven autonomy, they represent a powerful form of generative AI, marking a transition from reactive content generation into proactive multitasking capabilities. As an exemplar, we propose a...