RuleForge: Automated Generation and Validation for Web Vulnerability Detection at Scale

Security teams face a challenge: the volume of newly disclosed Common Vulnerabilities and Exposures CVEs far exceeds the capacity to manually develop detection mechanisms. In 2025, the National Vulnerability Database published over 48,000 new vulnerabilities, motivating the need for automation. W...

XMLDOM 安全漏洞

XMLDOM is a JavaScript implementation of the W3C DOM for Node developed by jindw. Versions of XMLDOM prior to 0.6.0, 0.8.12, and 0.9.9 contain security vulnerabilities. These vulnerabilities stem from allowing attackers to insert controlled strings into CDATASection nodes, potentially leading to...

Apple macOS 安全漏洞

Apple macOS Sequoia is an operating system from the American company Apple Apple. A code execution vulnerability exists in Apple macOS Sequoia, which is caused due to an error in the model I/O component when opening a specially crafted file, and can be exploited by an attacker to execute arbitrar...

AgentWatcher: A Rule-Based Prompt Injection Monitor

Large language models LLMs and their applications, such as agents, are highly vulnerable to prompt injection attacks. State-of-the-art prompt injection detection methods have the following limitations: 1 their effectiveness degrades significantly as context length increases, and 2 they lack...

Docker Desktop < 4.67.0 SSRF

The version of Docker Desktop is prior to 4.67.0. It is therefore affected by a server-side request forgery vulnerability. - Docker Model Runner contains an SSRF vulnerability in its OCI registry token exchange flow. When pulling a model, Model Runner follows the realm URL from the registry’s...

PT-2026-29951

Docker Model Runner OCI Registry Client Vulnerable to Server-Side Request Forgery SSRF in github.com/docker/model-runner...

Seclens: Role-Specific Evaluation of LLM'S for Security Vulnerablity Detection

Existing benchmarks for LLM-based vulnerability detection compress model performance into a single metric, which fails to reflect the distinct priorities of different stakeholders. For example, a CISO may emphasize high recall of critical vulnerabilities, an engineering leader may prioritize...

Apple macOS 安全漏洞

Apple macOS Sequoia is an operating system from the American company Apple Apple. Apple macOS Sequoia suffers from a code execution vulnerability that is caused due to an error in the model I/O component when opening a specially crafted file. An attacker can exploit the vulnerability to execute...

Docker Desktop < 4.67.0 SSRF (CVE-2026-33990)

The version of Docker Desktop installed on the remote host is prior to 4.67.0. It is, therefore, affected by a server-side request forgery SSRF vulnerability: - Docker Model Runner contains an SSRF vulnerability in its OCI registry token exchange flow. When pulling a model, Model Runner follows t...

vLLM 输入验证错误漏洞

vLLM is an open-source LLM-based inference and service engine that features high throughput and efficient memory usage. Versions of vLLM prior to 0.5.5 and 0.18.0 contained a vulnerability related to input validation errors. This vulnerability stemmed from inconsistencies in the audio mono downmi...

Synthetic Trust Attacks: Modeling How Generative AI Manipulates Human Decisions in Social Engineering Fraud

Imagine receiving a video call from your CFO, surrounded by colleagues, asking you to urgently authorise a confidential transfer. You comply. Every person on that call was fake, and you just lost $25 million. This is not a hypothetical. It happened in Hong Kong in January 2024, and it is becoming...

MCP Go SDK 安全漏洞

MCP Go SDK is an open-source development toolkit for the Model Context Protocol. Versions of MCP Go SDK prior to 1.4.0 contained security vulnerabilities. These vulnerabilities stemmed from the default lack of DNS rebinding protection, allowing malicious websites to bypass the same-origin policy...

PT-2026-29747

Name of the Vulnerable Software and Affected Versions Agno versions prior to 2.3.24 Description An arbitrary code execution issue exists in the model execution component. Attackers can execute arbitrary Python code by manipulating the field type parameter passed to the eval function. By influenci...

agno 安全漏洞

Agno is an open-source full-stack framework developed by Agno for building multi-agent systems with memory, knowledge, and reasoning capabilities. Versions of Agno prior to 2.3.24 contained a security vulnerability, which was caused by improper handling of the fieldtype parameter in the model...

fast-filesystem-mcp 命令注入漏洞

fast-filesystem-mcp is a model context protocol server developed by Efforthye. Versions of fast-filesystem-mcp 3.5.1 and earlier contained a command injection vulnerability. This vulnerability originated from the handleGetDiskUsage function in the file src/index.ts, which allowed command injectio...

AEGIS: Adversarial Entropy-Guided Immune System -- Thermodynamic State Space Models for Zero-Day Network Evasion Detection

As TLS 1.3 encryption limits traditional Deep Packet Inspection DPI, the security community has pivoted to Euclidean Transformer-based classifiers e.g., ET-BERT for encrypted traffic analysis. However, these models remain vulnerable to byte-level adversarial morphing -- recent pre-padding attacks...

The Quantum-Cryptographic Co-Evolution

As quantum computing matures toward the realization of Cryptographically Relevant Quantum Computers CRQC, global cryptographic infrastructure faces an existential threat. This paper introduces a two-dimensional coordinate system to map the co-evolution of cryptographic resilience x-axis and...

GHSA-W2FM-25VW-VH7F mcp-handler has a tool response leak across concurrent client sessions ('Race Condition')

mcp-handler versions prior to 1.1.0 accepted @modelcontextprotocol/sdk =1.26.0, which contains the fix for CVE-2026-25536. Workarounds - Upgrade @modelcontextprotocol/sdk to =1.26.0 note: the SDK will throw on transport reuse, which will break mcp-handler 1.1.0 which effectively forces the upgrad...

mcp-handler has a tool response leak across concurrent client sessions ('Race Condition')

mcp-handler versions prior to 1.1.0 accepted @modelcontextprotocol/sdk =1.26.0, which contains the fix for CVE-2026-25536. Workarounds - Upgrade @modelcontextprotocol/sdk to =1.26.0 note: the SDK will throw on transport reuse, which will break mcp-handler 1.1.0 which effectively forces the upgrad...

GHSA-324Q-CWX9-7CRR KubeAI: OS Command Injection via Model URL in Ollama Engine startup probe allows arbitrary command execution in model pods

CHAMP: Description Summary The ollamaStartupProbeScript function in internal/modelcontroller/engineollama.go constructs a shell command string using fmt.Sprintf with unsanitized model URL components ref, modelParam. This shell command is executed via bash -c as a Kubernetes startup probe. An...