Sentinel-GPT

Sentinel-GPT: Agentic AI for Threat Hunting 🛡️ An automated cybersecurity pipeline that transforms raw Azure logs into actionable intelligence using LLM-based reasoning. 📖 Overview Sentinel-GPT is an AI Agent specialised for threat hunting in cybersecurity. It is integrated with Azure Log Analytic...

CVE-2026-34724

🚨 CVE-2026-34724 - Zammad 7.0.1 - Server-Side Template Injection SSTI to Remote Code Execution RCE via AI Agent 📌 Overview Zammad is a popular open-source web-based helpdesk and customer support system. CVE-2026-34724 is a Server-Side Template Injection SSTI vulnerability in the new AI Agent...

CVE-2026-88938 knowns through 0.33.0 Path Traversal via code.find MCP tool

knowns through 0.33.0 fails to confine the path argument of the code.find MCP tool to the project root, allowing AI agent sessions to read source files anywhere on the host. Attackers can supply absolute paths or relative traversal sequences to the path argument and retrieve full file contents fr...

CVE-2026-88938 knowns through 0.33.0 Path Traversal via code.find MCP tool

knowns through 0.33.0 fails to confine the path argument of the code.find MCP tool to the project root, allowing AI agent sessions to read source files anywhere on the host. Attackers can supply absolute paths or relative traversal sequences to the path argument and retrieve full file contents fr...

CVE-2026-88938

knowns through version 0.33.0 is affected by a path traversal vulnerability in the code.find MCP tool . The tool fails to confine its path argument to the project root, allowing an AI agent session to supply absolute paths or relative traversal sequences and retrieve full file contents from outsi...

CVE-2026-88938 knowns through 0.33.0 Path Traversal via code.find MCP tool

knowns through 0.33.0 fails to confine the path argument of the code.find MCP tool to the project root, allowing AI agent sessions to read source files anywhere on the host. Attackers can supply absolute paths or relative traversal sequences to the path argument and retrieve full file contents fr...

geiger

geiger Un contador Geiger para agentes de IA. Un único comando de solo lectura que inventaría cada agente de IA, harness, servidor MCP, plugin y extensión de IA en una máquina — y te dice, en lenguaje sencillo, a qué puede acceder cada uno. root@kitploit: npx geiger-scan Sin instalación. Sin...

React19-fix-vibecoders

React 19 y Next.js: prompts de corrección de seguridad para agentes de IA 🚨 VULNERABILIDAD DE SEGURIDAD CRÍTICA - CVE-2025-55182 y CVE-2025-66478 Este repositorio contiene prompts reutilizables para ayudar a los asistentes de codificación con IA a corregir automáticamente vulnerabilidades de...

CVE-2025-13374

Kalrav AI Agent = 2.3.3 - Unauthenticated Arbitrary File Upload via kalravuploadfile AJAX Action The WordPress Kalrav AI Agent plugin version 2.3.3 and prior contains an arbitrary file upload vulnerability that allows unauthenticated users to upload malicious PHP files to the server, potentially...

Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6

Anthropic on Wednesday disclosed a fourth incident in which its artificial intelligence AI model broke into real third-party systems, marking the latest in a growing list of cases that have raised concerns about the security risks posed by autonomous AI agents. The AI company said the incident...

agentZ

AgentZ Plataforma de IA Agéntica de Confianza Cero Construye, ejecuta y gobierna agentes de IA. Seguro por diseño. Cada agente se ejecuta dentro de un sandbox de denegación por defecto y nunca recibe una credencial. Cada acción del agente es trazable. AgentZ es una plataforma multiinquilino para...

demo-cve-2026-4821

Demo: Reparación de la vulnerabilidad CVE-2026-4821 Este repositorio demuestra cómo el sistema de procedencia de OpenExecution rastrea la reparación de vulnerabilidades por parte de agentes de IA con total auditabilidad criptográfica. Escenario Vulnerabilidad : Inyección SQL en el módulo de...

CVE-2026-54520

CVE-2026-54520 El path traversal en el paso de archivos del flujo de trabajo de ai-agent-automation permitía la lectura/escritura fuera del directorio esperado. Aviso principal: GHSA-cm8g-8jfq-887p Crédito al investigador: @chaitanyagarware De un vistazo...

CVE-2026-54519

CVE-2026-54519 ai-agent-automation 在 memory APIs 中缺少所有权检查，允许跨用户的内存读取和删除。 主要公告： GHSA-qv97-83w4-ff86 研究人员致谢： @chaitanyagarware 概览 字段| 值 ---|--- CVE| CVE-2026-54519 GHSA| GHSA-qv97-83w4-ff86 项目| vmDeshpande/ai-agent-automation 包| backend 生态系统| npm 严重性| 高 CVSS| 8.8,...

sdk

@cosignet/sdk 轻量、零依赖的 TypeScript 客户端，用于 Cosignet — 人在回路中 批准高风险 AI 代理操作， 并附带绑定负载的通行密钥签名。 在危险操作执行前引入人工审批：暂停操作，获取一个人的明确通行密钥批准（Face ID / Touch ID / Windows Hello / 安全密钥），仅在签署的决策 绑定到确切负载 后才继续执行——事后修改操作则签名不再匹配。Cosignet 是一个审批和证据层，而非执行器或策略引擎。可在任何支持全局 fetch + Web Crypto 的环境中运行：Node 18+、Cloudflare...

Muse, Meta’s New Personal AI Agent, Needs You to Trust It

Designed to compete with OpenClaw and Instinct, the company says Muse can do everything from sell your car to book you a plane ticket...

clawguard

🦞 ClawGuard v3 Kit de Seguridad Empresarial para Agentes de IA - Defensa Activa Impulsada por SKILL.md Concepto Principal ¡La defensa principal de ClawGuard v3 no está en el código, sino en SKILL.md! El SKILL.md de cada módulo es en sí mismo una guía de defensa completa: Le dice al Agente cuándo...

third-thoughts

Terceros Pensamientos Un proyecto de investigación de Lightless Labs que estudia el comportamiento de los agentes de IA a escala mediante análisis multidisciplinario de corpus básicamente, lanzando cosas contra la pared y viendo qué se queda pegado. Nombrado a partir del concepto de Tiffany Achin...

masq

Sit where the model sits. See what it can see. MASQ MASQ is a local security CLI for AI agent stacks. It sits in the same seat the model sits in: it reads MCP tool catalogs, completes the MCP handshake, looks through session logs, and probes local model HTTP ports. Then it prints findings and...

node9-proxy v2.8.3

🛡️ Node9 What did your AI agent actually do? Find out. Node9 sits between your AI agent and the tools it can use — discover what it's already been doing, protect against risky actions in real time, and review what happened over any time window. Works with Claude Code · Codex CLI · Antigravity agy ...