OpenCode < 1.0.216 - Unauthenticated Remote Code Execution

OpenCode versions prior to 1.0.216 contain an unauthenticated remote code execution vulnerability. The application exposes session and shell execution endpoints without proper authentication, allowing remote attackers to create sessions and execute arbitrary shell commands on the underlying serve...

opencode

开源的 AI Coding Agent。 English | 简体中文 | 繁體中文 | 한국어 | Deutsch | Español | Français | Italiano | Dansk | 日本語 | Polski | Русский | Bosanski | العربية | Norsk | Português Brasil | ไทย | Türkçe | Українська | বাংলা | Ελληνικά | Tiếng Việt 安装 root@kitploit: 直接安装 YOLO curl -fsSL https://opencode.ai/instal...

CVE-2026-22812

🔐 CVE-2026-22812 - Servidor HTTP no autenticado de OpenCode → Ejecución Remota de Código RCE Aviso de seguridad, análisis técnico, guía de detección y notas de mitigación para CVE-2026-22812 📌 Resumen CVE-2026-22812 es una vulnerabilidad de Ejecución Remota de Código RCE de alta gravedad que afec...

CVE-2026-22812

CVE-2026-22812 CVE-2026-22812 - Plantilla de Nuclei para RCE no autenticado en OpenCode para escaneo. solicitud de burp suite root@kitploit: POST /session HTTP/1.1 Host: target.com User-Agent: Mozilla/5.0 X11; Linux x8664 Content-Type: application/json Content-Length: 2 root@kitploit: POST...

opencode-secure

El agente de codificación de IA de código abierto. Instalación root@kitploit: YOLO curl -fsSL https://opencode.ai/install | bash Gestores de paquetes npm i -g opencode-ai@latest o bun/pnpm/yarn scoop install opencode Windows choco install opencode Windows brew install anomalyco/tap/opencode macOS...

CVE-2026-22812-OpenCode-RCE-exp

CVE-2026-22812-OpenCode-RCE-exp Método de uso: java -jar xxx.jar Entorno recomendado: jdk8...

CVE-2026-22812-exploit

Herramienta de Explotación CVE-2026-22812 Kit de explotación completo para CVE-2026-22812 - RCE no autenticado en OpenCode 🎯 Descripción general La Herramienta de Explotación CVE-2026-22812 es un framework de explotación integral para la vulnerabilidad de ejecución remota de código no autenticada...

CVE-2026-22812

CVE-2026-22812 Descripción general Una herramienta de explotación en Python para la vulnerabilidad de Ejecución Remota de Código RCE en OpenCode que afecta a versiones v1.0.216. Esta herramienta proporciona múltiples vectores de ataque, incluida la ejecución de comandos, lectura/escritura de...

CVE-2026-22812-POC

Explotador OpenCode CVE-2026-22812 Herramientas Python simples para verificar y explotar la vulnerabilidad de OpenCode CVE-2026-22812. Instalación root@kitploit: git clone https://github.com/CayberMods/CVE-2026-22812-POC.git cd CVE-2026-22812-POC pip install requests urllib3 argparse...

CVE-2026-88899

CVE-2026-88899 - Knowns 0.30.0 Inyección de Encabezados No Autenticada Otorga al Agente de IA Acceso Sin Restricciones al Sistema de Archivos del Host Descripción General Se reportó una vulnerabilidad de autorización indebida y suplantación de encabezados relacionada con el proxy inverso de Known...

CVE-2026-88899 knowns before 0.31.0 External Control of Agent Working Directory via x-opencode-directory Header

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. Remote attackers can supply arbitrary directory paths to execute file operations outside the project root on the host system...

EUVD-2026-75579

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. Remote attackers can supply arbitrary directory paths to execute file operations outside the project root on the host system...

CVE-2026-88899 knowns before 0.31.0 External Control of Agent Working Directory via x-opencode-directory Header

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. Remote attackers can supply arbitrary directory paths to execute file operations outside the project root on the host system...

CVE-2026-88899 knowns before 0.31.0 External Control of Agent Working Directory via x-opencode-directory Header

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. Remote attackers can supply arbitrary directory paths to execute file operations outside the project root on the host system...

CVE-2026-88899

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. This allows remote, unauthenticated attackers to supply arbitrary directory paths, enabling file operations outside the project root on the host system. The vulnera...

PT-2026-89477

knowns versions before 0.31.0 fail to properly validate the x-opencode-directory request header in the /api/opencode proxy endpoint. Remote attackers can supply arbitrary directory paths to execute file operations outside the project root on the host system...

opencode-1.18.26-1.1 on GA media (moderate)

opencode-1.18.26-1.1 on GA media Announcement ID: openSUSE-SU-2026:11673-1 Rating: moderate Cross-References: CVE-2026-23736 CVE-2026-23737 CVE-2026-59940 Affected Products: openSUSE Tumbleweed An update that solves 3 vulnerabilities can now be installed. Description: These are all security issue...

OpenCode Unauthenticated Remote Code Execution

OpenCode versions prior to 1.0.216 contain an unauthenticated remote code execution vulnerability that can allow a remote attacker to obtain a session identifier and execute arbitrary commands through the session shell endpoint. This proof of concept retrieves an available session and demonstrate...

OPENSUSE-SU-2026:11673-1 opencode-1.18.26-1.1 on GA media

These are all security issues fixed in the opencode-1.18.26-1.1 package on the GA media of openSUSE Tumbleweed...

AWS, Google, and Vercel Agent Flaws Let Attackers Trigger Tools Without Running the Model

Security flaws in agent infrastructure from Amazon Web Services AWS, Google, and Vercel let untrusted or forged instructions reach an agent's tools with no check that a model turn had authorized them. In several of the attack paths, the model never ran at all, so system prompts, content filters,...