CVE-2026-4593

CVE-2026-4593 describes a SQL injection in the EruptDataQuery function (erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java) within the MCP Tool Interface of erupts erupt bis 1.13.3. The issue arises from a manipulation that enables remote exploitation via a crafted input, with the ...

CVE-2026-4593 erupts erupt MCP Tool EruptDataQuery.java EruptDataQuery sql injection

A flaw has been found in erupts erupt bis 1.13.3. Affected by this vulnerability is the function EruptDataQuery of the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java of the component MCP Tool Interface. This manipulation causes sql injection hibernate. It is possible to...

CVE-2026-4593 erupts erupt MCP Tool EruptDataQuery.java EruptDataQuery sql injection

A flaw has been found in erupts erupt bis 1.13.3. Affected by this vulnerability is the function EruptDataQuery of the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java of the component MCP Tool Interface. This manipulation causes sql injection hibernate. It is possible to...

gimp: GIMP: Remote Code Execution via uninitialized memory in PGM file parsing

A flaw was found in GIMP. This uninitialized memory vulnerability allows a remote attacker to execute arbitrary code on affected installations. Successful exploitation requires user interaction, where the target must open a specially crafted PGM Portable Graymap image file. This can lead to...

ERUPT 安全漏洞

ERUPT is a low-code + AI-based framework developed by YuePeng, a personal developer in China. Version 1.13.3 of ERUPT contains a security vulnerability. This vulnerability stems from incorrect operations on the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java in the MCP Tool...

Model Context Protocol Threat Modeling and Analyzing Vulnerabilities to Prompt Injection with Tool Poisoning

The Model Context Protocol MCP has rapidly emerged as a universal standard for connecting AI assistants to external tools and data sources. While MCP simplifies integration between AI applications and various services, it introduces significant security vulnerabilities, particularly on the client...

PT-2026-27150

A flaw has been found in erupts erupt bis 1.13.3. Affected by this vulnerability is the function EruptDataQuery of the file erupt-ai/src/main/java/xyz/erupt/ai/call/impl/EruptDataQuery.java of the component MCP Tool Interface. This manipulation causes sql injection hibernate. It is possible to...

AI in Cybersecurity Education -- Scalable Agentic CTF Design Principles and Educational Outcomes

Large language models are rapidly changing how learners acquire and demonstrate cybersecurity skills. However, when human--AI collaboration is allowed, educators still lack validated competition designs and evaluation practices that remain fair and evidence-based. This paper presents a...

Auditing MCP Servers for Over-Privileged Tool Capabilities

The Model Context Protocol MCP has emerged as a standard for connecting Large Language Models LLMs to external tools and data. However, MCP servers often expose privileged capabilities, such as file system access, network requests, and command execution that can be exploited if not properly...

CVE-2026-32898

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

CVE-2026-32898 OpenClaw < 2026.2.23 - ACP Permission Auto-Approval Bypass via Untrusted Tool Metadata

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

EUVD-2026-13976

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

CVE-2026-32898 OpenClaw < 2026.2.23 - ACP Permission Auto-Approval Bypass via Untrusted Tool Metadata

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

CVE-2026-32898

CVE-2026-32898 affects OpenClaw versions prior to 2026.2.23, specifically the ACP client. The issue is an authorization bypass where tool calls are auto-approved based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-cl...

CVE-2026-32051 OpenClaw < 2026.3.1 - Authorization Bypass in Agent Runs via Owner-Only Tool Access

OpenClaw versions prior to 2026.3.1 contain an authorization mismatch vulnerability that allows authenticated callers with operator.write scope to invoke owner-only tool surfaces including gateway and cron through agent runs in scoped-token deployments. Attackers with write-scope access can perfo...

PT-2026-26747

OpenClaw versions prior to 2026.2.23 contain an authorization bypass vulnerability in the ACP client that auto-approves tool calls based on untrusted toolCall.kind metadata and permissive name heuristics. Attackers can bypass interactive approval prompts for read-class operations by spoofing tool...

OpenClaw 安全漏洞

OpenClaw is an intelligent artificial assistant open-sourced by OpenClaw. OpenClaw suffers from a security vulnerability that can be exploited by an attacker to cause an authenticated caller with operator.write scope to invoke the owner-only tool interface...

T-MAP: Red-Teaming LLM Agents with Trajectory-Aware Evolutionary Search

While prior red-teaming efforts have focused on eliciting harmful text outputs from large language models LLMs, such approaches fail to capture agent-specific vulnerabilities that emerge through multi-step tool execution, particularly in rapidly growing ecosystems such as the Model Context Protoc...

CVE-2026-33147

GMT is an open source collection of command-line tools for manipulating geographic and Cartesian data sets. In versions from 6.6.0 and prior, a stack-based buffer overflow vulnerability was identified in the gmtremotedatasetid function within src/gmtremote.c. This issue occurs when a specially...

CVE-2026-33144

GPAC is an open-source multimedia framework. Prior to commit 86b0e36, a heap-based buffer overflow write vulnerability was discovered in GPAC MP4Box. The vulnerability exists in the gfxmlparsebitsequencebs function in utils/xmlbincustom.c when processing a crafted NHML file containing malicious...