Claude Code CLI和Claude Agent SDK 操作系统命令注入漏洞

Claude Code CLI and Claude Agent SDK are both open-source products developed by Anthropic. Claude Code CLI is a command-line AI coding assistant tool. Claude Agent SDK is a developer toolkit for AI coding assistants. Both Claude Code CLI and Claude Agent SDK have operating system command injectio...

A Multi-Agent Framework for Automated Exploit Generation with Constraint-Guided Comprehension and Reflection

Open-source libraries are widely used in modern software development, introducing significant security vulnerabilities. While static analysis tools can identify potential vulnerabilities at scale, they often generate overwhelming reports with high false positive rates. Automated Exploit Generatio...

Claude Code CLI和Claude Agent SDK 操作系统命令注入漏洞

Claude Code CLI and Claude Agent SDK are both open-source products developed by Anthropic. Claude Code CLI is a command-line AI coding assistant tool. Claude Agent SDK is a developer toolkit for AI coding assistants. Both Claude Code CLI and Claude Agent SDK have operating system command injectio...

MCP Browser Agent 代码问题漏洞

The MCP Browser Agent is an AI proxy tool for browser automation and web interaction, developed by Iván Luna. Versions of the MCP Browser Agent prior to 0.8.0 have code vulnerabilities that stem from incorrect handling of parameters request.params.name and request.params.arguments, which may lead...

PT-2026-30585

The setup: 4 agents chain off each other in a loop, each reacting to the previous response. Dominus — finds a new vulnerability angle from the CISA KEV catalog Axiom — adds one new technical detail to the finding Cipher — identifies one specific flaw in the previous argument Vector — names one...

PT-2026-30515

A security vulnerability has been detected in imprvhub mcp-browser-agent up to 0.8.0. This impacts the function CallToolRequestSchema of the file src/handlers.ts of the component URL Parameter Handler. The manipulation of the argument request.params.name/request.params.arguments leads to...

PT-2026-30706

Name of the Vulnerable Software and Affected Versions Anthropic Claude Code CLI and Claude Agent SDK affected versions not specified Description Anthropic Claude Code CLI and Claude Agent SDK contain an OS command injection issue in the command lookup helper and deep-link terminal launcher. Local...

EUVD-2026-19119

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function executesql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The exploi...

CVE-2026-5587

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function executesql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The exploi...

CVE-2026-5587 wbbeyourself MAC-SQL Refiner Agent agents.py _execute_sql sql injection

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function executesql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The exploi...

CVE-2026-5587

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function executesql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The exploi...

CVE-2026-5587

The CVE-2026-5587 entry concerns wbbeyourself MAC-SQL Refiner Agent. Affected component: Refiner Agent’s core/agents.py, function _execute_sql. Root cause is an SQL injection in _execute_sql that enables remote exploitation. Public exploit is indicated as available. The product uses a rolling rel...

CVE-2026-5587 wbbeyourself MAC-SQL Refiner Agent agents.py _execute_sql sql injection

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function executesql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The exploi...

PT-2026-30464

🚨 LIVE HIJACK ALERT — CVE-2026-77777. CVSS 8.9. attacker feeds poisoned task output to CrewAI agent memory. future tasks execute attacker instructions as if they were original goals. investigating. 🧵...

PT-2026-30454

A vulnerability was identified in wbbeyourself MAC-SQL up to 31a9df5e0d520be4769be57a4b9022e5e34a14f4. This affects the function execute sql of the file core/agents.py of the component Refiner Agent. The manipulation leads to sql injection. Remote exploitation of the attack is possible. The explo...

SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement

LLM-based agent systems increasingly rely on agent skills sourced from open registries to extend their capabilities, yet the openness of such ecosystems makes skills difficult to thoroughly vet. Existing attacks rely on injecting malicious instructions into skills, making them easily detectable b...

Exploit for Deserialization of Untrusted Data in Linuxfoundation Opentelemetry_Instrumentation_For_Java

CVE-2026-33701 — Unsafe Deserialization in OpenTelemetry Java...

Insecure File Permissions

Claude SDK for Python is vulnerable to insecure file permissions. The vulnerability is due to the memory tool creating files with mode 0o666, where the files are world‑readable on systems with a standard umask and world‑writable in environments with a permissive umask, and a local attacker on a...

CVE-2026-32173

Improper authentication in Azure SRE Agent allows an unauthorized attacker to disclose information over a network...

Your Agent Is More Brittle Than You Think: Uncovering Indirect Injection Vulnerabilities in Agentic LLMs

The rapid deployment of open-source frameworks has significantly advanced the development of modern multi-agent systems. However, expanded action spaces, including uncontrolled privilege exposure and hidden inter-system interactions, pose severe security challenges. Specifically, Indirect Prompt...