CVE-2026-105740

Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to 1.9.0, any authenticated Langflow user can achieve Remote Code Execution RCE on the server by adding an MCP server with the "Stdio" transport. The user-supplied command field is passed directly to bash -c "exe...

EUVD-2026-92799

Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to 1.9.0, any authenticated Langflow user can achieve Remote Code Execution RCE on the server by adding an MCP server with the "Stdio" transport. The user-supplied command field is passed directly to bash -c "exe...

mcp-attack-detection-sentinel

MCP 攻击检测 — Sentinel 研究实验室 针对 MCP（模型上下文协议）攻击模式 的 Sentinel 检测研究：MCP 相关事件后的异常身份访问、工具定义篡改、跨资源访问，以及可能的利用后行为。这些规则映射到 OWASP 2026 年智能体应用十大风险。 发布状态： 此仓库目前没有已发布的 Nine Lives 配套文章，并且有意未列入网站的 Labs 目录。它仍然是一个独立的研究项目。 CVE-2026-26118 规则是一种基于行为的启发式规则，参考了 Azure MCP Server SSRF 披露。它不 证明该 CVE 已被利用或令牌已被窃取。微软将该问题描述为...

mcpjam-cve-2026-23744-validator

MCPJam Inspector 授权安全验证器 一个针对影响易受攻击 MCPJam Inspector 部署的 CVE-2026-23744 的有界概念验证验证器。 该项目是一个独立的、非破坏性的验证工具，用于先前披露的 CVE-2026-23744。它不是原始漏洞披露或漏洞利用。 此仓库旨在供安全专业人员在验证他们拥有或已获得明确书面授权的系统时使用。 重要提示 此验证器通过使目标 MCPJam Inspector 进程执行一个固定的、非交互式的证据命令来确认暴露风险。它不提供通用的漏洞利用框架、shell、后利用工作流、持久化机制、凭据访问例程或权限提升能力。...

CVE-2026-89039

A caller who can invoke the convertplaywrightscript prompt in mcp-k6 can pass a bare file path as the playwrightscript argument and receive the contents of any file readable by the user running the server, including SSH keys and cloud credentials in that user's home directory path traversal. The...

CVE-2026-89039 Arbitrary file read via the convert_playwright_script prompt in mcp-k6

A caller who can invoke the convertplaywrightscript prompt in mcp-k6 can pass a bare file path as the playwrightscript argument and receive the contents of any file readable by the user running the server, including SSH keys and cloud credentials in that user's home directory path traversal. The...

CVE-2026-89039 Arbitrary file read via the convert_playwright_script prompt in mcp-k6

A caller who can invoke the convertplaywrightscript prompt in mcp-k6 can pass a bare file path as the playwrightscript argument and receive the contents of any file readable by the user running the server, including SSH keys and cloud credentials in that user's home directory path traversal. The...

CVE-2026-89039

The k6 MCP server is vulnerable to a path traversal flaw in the convert_playwright_script prompt via the playwright_script argument. While @-prefixed paths are restricted to the current working directory, bare paths follow an undocumented code path with no restrictions, allowing callers to read a...

EUVD-2026-92434

A caller who can invoke the convertplaywrightscript prompt in mcp-k6 can pass a bare file path as the playwrightscript argument and receive the contents of any file readable by the user running the server, including SSH keys and cloud credentials in that user's home directory path traversal. The...

ai-kubernetes-helm-chart-security

IA en Kubernetes: valores predeterminados de seguridad de los Helm charts evidencia Este repositorio contiene los datos y la evidencia que respaldan el informe técnico de Sorami "AI on Kubernetes: 10 of 14 Helm charts ship with no API auth" versión 1.2. Lea el informe aquí:...

super-trouper

Super Trouper Un servidor MCP para el kit de herramientas de ingeniería inversa Frida. Super Trouper ejecuta un servidor MCP sobre STDIO que expone las capacidades de Frida como herramientas para agentes de programación. Puede conectarse a dispositivos locales, USB y remotos; inspeccionar...

PT-2026-106235

Name of the Vulnerable Software and Affected Versions Langflow versions prior to 1.9.0 Description Authenticated users can achieve Remote Code Execution RCE on the server by adding a Model Context Protocol MCP server using the "Stdio" transport. The command field is passed directly to bash -c "ex...

PT-2026-106273

Name of the Vulnerable Software and Affected Versions Dify versions prior to 1.16.0 Description An authenticated workspace member can modify the status and parameters of another application's MCP server, which could lead to service disruption or data redirection. This occurs because the...

PT-2026-106004

The convert playwright script prompt in the k6 MCP server accepts a file path as its playwright script argument. Paths given in the documented '@'-prefixed form are restricted to the server's current working directory, but a bare path is resolved by a separate undocumented code path that applies ...

ida-pro-mcp

IDA Pro MCP ⚠️ Archived — superseded by the official Hex-Rays IDA MCP server This project is no longer maintained. On 2026-09-27 Hex-Rays SA released the Official IDA MCP Server and IDA Nexus. Development stopped the same day at 1.0.0a4. For current use, install theirs: uvx ida-hcli mcp install. I...

claude-project-scanner

Claude 项目扫描器 一种只读的 Claude Code 技能，用于在打开 第三方项目之前扫描其已知安全风险。 检测到以下文档中记载的攻击模式： CVE-2025-59536（Claude Code 生命周期钩子注入） CVE-2025-61260（OpenAI Codex CODEXHOME 注入） CVE-2025-54136（Cursor 插件验证绕过） 此外还检测 statusLine.command 执行、additionalDirectories 逃逸、代理/技能/命令/输出样式中的提示注入、可疑的 MCP 服务器以及隐藏的 base64 形式载荷。 ⚠️ 请先阅读 1...

ctxdebug

MCP servers for WinDbg, IDA Pro, and x64dbg One stdio interface · 160+ tools · Three debuggers, one client. Quick Start · Architecture · Servers · Workflows · Report an Issue Website · ctxdebug.xyz · Contact · [email protected] · Status · Public alpha For authorized reverse engineering, cras...

CVE-2026-104120

A security vulnerability has been detected in modelcontextprotocol mcp-server-fetch and mcp-server-everything up to 2026.6.4. Affected is the function fetchurl of the file mcpserverfetch/server.py of the component Fetch Tool. The manipulation of the argument url/path leads to server-side request...

CVE-2026-104120 modelcontextprotocol mcp-server-fetch/mcp-server-everything Fetch Tool server.py fetch_url server-side request forgery

A security vulnerability has been detected in modelcontextprotocol mcp-server-fetch and mcp-server-everything up to 2026.6.4. Affected is the function fetchurl of the file mcpserverfetch/server.py of the component Fetch Tool. The manipulation of the argument url/path leads to server-side request...

CVE-2026-104120

A server-side request forgery (SSRF) vulnerability exists in the modelcontextprotocol components mcp-server-fetch and mcp-server-everything up to version 2026.6.4 . The flaw is located in the fetch_url function within the file mcp_server_fetch/server.py of the Fetch Tool component. An attacker ca...