CVE-2026-59971

MySQL MCP Server is a Model Context Protocol server that enables secure interaction with MySQL databases. Prior to 0.4.2, setting MCPTRANSPORT=sse causes src/mysqlmcpserver/server.py to construct SseServerTransport without securitysettings or enablednsrebindingprotection, while the Starlette rout...

MCP-Inspector-CVE-2025-49596

MCP Inspector CVE-2025-49596 脆弱性チェッカー このスクリプトは、MCP Inspector サーバーが CVE-2025-49596 に対して脆弱かどうかをチェックします。これは、/sse エンドポイントが認証されていないリクエスト（認証トークンがない）に応答するというセキュリティ問題です。 動作の仕組み スクリプトは、対象の MCP Inspector サーバーの /sse エンドポイントに GET リクエストを送信します。 応答をチェックします: サーバーが認証トークンを必要とせずに HTTP 200 OK で応答した場合、脆弱である可能性が高いです。...

damn-vulnerable-MCP-server

Damn Vulnerable MCP の Cursor SSE 互換性 このドキュメントでは、Cursor の Server-Sent Events SSE 実装と互換性を持たせるために Damn Vulnerable Model Context Protocol DVMCP プロジェクトに加えられた更新内容の概要を説明します。 概要 元の DVMCP プロジェクトは、Cursor の SSE 実装をサポートするように拡張され、チャレンジサーバーに Cursor から直接アクセスできるようになりました。これにより、実際の MCP...

sandbox-agent

코딩 에이전트를 샌드박스에서 실행하고 HTTP로 제어하세요 샌드박스 내부에서 실행되는 서버입니다. 앱이 원격으로 연결하여 Claude Code, Codex, OpenCode, Cursor, Amp, Pi를 제어하며 이벤트 스트리밍, 권한 처리, 세션 관리를 수행합니다. 문서 — API 레퍼런스 — Discord 실험적: Gigacode — 모든 코딩 에이전트와 함께 OpenCode의 TUI를 사용합니다. Sandbox Agent가 필요한 이유 코딩 에이전트를 원격으로 실행하는 것은 어렵습니다. 기존 SDK는 로컬 실행을 가정하...

studio

ShipSec Studio 오픈소스 보안 워크플로 오케스트레이션 플랫폼. ShipSec은 현재 활발히 개발 중입니다. 안정적인 프로덕션 사용과 고성능 보안 운영을 위해 플랫폼을 최적화하고 있습니다. ShipSec Studio는 자동화된 보안 워크플로를 구축, 실행 및 모니터링하기 위한 시각적 DSL과 런타임을 제공합니다. 보안 로직을 인프라 관리에서 분리하여, 대규모 보안 도구 실행을 위한 내구성 있고 격리된 환경을 제공합니다. YouTube에서 플랫폼 동작을 확인하세요. 🏗️ 핵심 기둥 내구성 있는 재개 가능 워크플로 —...

firecrawl-mcp-server

Firecrawl MCP 서버 MCP 호환 AI 에이전트에 Firecrawl을 제공하는 Model Context Protocol MCP 서버입니다. — 깨끗하고 에이전트에 적합한 컨텍스트를 위해 실시간 웹을 검색, 스크랩, 상호작용할 수 있습니다. 초기 구현에 기여해 주신 @vrknetha, @knacklabs 님께 큰 감사를 드립니다! 기능 웹 검색 및 전체 페이지 콘텐츠 가져오기 모든 URL을 깨끗하고 구조화된 데이터로 스크랩 페이지와 상호작용 — 클릭, 탐색, 조작 자율 에이전트를 이용한 심층 연구 자동 재시도 및 속도 ...

pentest-mcp

Pentest MCP 最新のトランスポート/認証サポートと拡張された偵察ツールを備えた、プロフェッショナルな侵入テスト用 MCP サーバー。 0.9.0 の変更点 MCP SDK を @modelcontextprotocol/sdk@^1.26.0 にアップグレード MCP Inspector は最新リリース @modelcontextprotocol/inspector@^0.20.0 のバンドルランチャーを維持 Streamable HTTP がプライマリネットワークトランスポートに MCPTRANSPORT=http SSE は非推奨の互換モードとしてのみ利用可能 Bearer...

CVE-2026-57128

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.58, the SSE server in src/praisonai-agents/praisonaiagents/server/server.py does not consult ServerConfig.authtoken before handling /publish, /events, or /info requests. A network client that can reach the server can broadcast...

CVE-2026-57123

PraisonAI (praisonaiagents), a multi-agent teams system, is affected in its MCP SSE transport (ToolsMCPServer.run_sse() / launch_tools_mcp_server(transport="sse")). The server binds to 0.0.0.0 by default and builds its Starlette app without authentication middleware or Origin-header validation. T...

CVE-2026-57123 PraisonAI: MCP SSE transport binds 0.0.0.0 with no authentication and no Origin validation; bundled SecurityConfig is never wired in

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.59, ToolsMCPServer.runsse and launchtoolsmcpserver bind to 0.0.0.0 and create /sse and /messages/ routes without invoking the available SecurityConfig authentication, origin-validation, or DNS-rebinding controls. Any reachable...

CVE-2026-57128

PraisonAI (praisonaiagents

CVE-2026-57128 PraisonAI: Unauthenticated Event Injection via SSE `/publish` Endpoint

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.58, the SSE server in src/praisonai-agents/praisonaiagents/server/server.py does not consult ServerConfig.authtoken before handling /publish, /events, or /info requests. A network client that can reach the server can broadcast...

MySQL MCP Server: Missing Origin/Host Validation in SSE Transport Enables Unauthenticated SQL Execution (DNS Rebinding / Direct Exposure)

Summary In SSE/HTTP transport mode, mysqlmcpserver constructs SseServerTransport without passing securitysettings. As a result, the MCP Python SDK's DNS-rebinding protection Origin/Host header validation is disabled; the Starlette application has no CORS or TrustedHost middleware; and the service...

CVE-2026-34227

未認証のMCPインターフェース + CORSバイパス SliverのMCPサーバーは未認証のSSEインターフェース を起動し、すべてのレスポンスにAccess-Control-Allow-Origin: を返します。 mcp-goライブラリはContent-Typeを検証しないため、シンプルリクエスト （text/plain）はブラウザのプリフライトチェックを完全にスキップします。OPTIONSなし。CORSブロックなし。認証なし。オペレーターが訪問した任意のページが、そのC2と静かに通信できます。 → GitHubセキュリティアドバイザリ GHSA-6fpf-248c-m7wm...

GHSA-F2FP-RGF2-35CP HTTPX2: Quadratic SSE line buffering can cause CPU denial of service

Summary HTTPX2's Server-Sent Events SSE parser repeatedly copied and rescanned buffered text when a server split one unterminated line across many response chunks. The total work grows quadratically with the length of the line. An attacker-controlled or compromised SSE endpoint can exploit this...

EUVD-2026-70242

HTTPX2: Quadratic SSE line buffering can cause CPU denial of service...

Stream Corruption

Spring Framework is vulnerable to stream corruption. The vulnerability is due to improper handling of response streams when processing SSE with view fragments, which allows an attacker to potentially corrupt or disrupt the application’s HTTP response stream...

CVE-2026-86252

h3 versions before 1.15.9 fail to sanitize carriage return characters in EventStream data and comment fields, allowing attackers to inject arbitrary SSE events by including unsanitized carriage returns. Attackers can inject event type directives, split single push calls into multiple browser-pars...

CVE-2026-86252

h3 versions before 1.15.9 fail to sanitize carriage return characters in EventStream data and comment fields, enabling attackers to inject arbitrary SSE events . The flaw allows injection of event type directives, splitting a single push call into multiple browser-parsed events, or escaping comme...

CVE-2026-86252 h3 before 1.15.9 SSE Event Injection via Carriage Return

h3 versions before 1.15.9 fail to sanitize carriage return characters in EventStream data and comment fields, allowing attackers to inject arbitrary SSE events by including unsanitized carriage returns. Attackers can inject event type directives, split single push calls into multiple browser-pars...