Skip to content
CVSS 10.0CVSS 10.0 · CRITICAL

CVE-2026-73678RCE sin autenticación en MindsDB Minds Platform

MindsDB Minds Platform version 26.1.0 and earlier contains an unauthenticated remote code execution vulnerability that allows unauthenticated attackers to execute arbitrary OS commands by submitting crafted prompts to the unprotected POST /api/v1/responses/ endpoint, which reaches the Anton agent's scratchpad tool that calls exec() on attacker-influenced Python source without sandboxing. Attackers can first configure their own LLM API key through the unauthenticated PUT /api/v1/settings/ endpoint, then POST a prompt directing the agent to invoke the scratchpad tool with arbitrary Python code, achieving full OS command execution as the user running the desktop application and enabling access to SSH keys, stored credentials, and environment secrets.

Ver en NVD

Análisis

MindsDB permite la ejecución remota de código sin autenticación a través de una vulnerabilidad en el agente Anton que ejecuta código Python sin aislamiento. Un atacante puede tomar control total del sistema operativo y acceder a llaves SSH, credenciales y secretos del entorno mediante prompts maliciosos. Esta falla crítica compromete la infraestructura que aloja modelos de IA y flujos de datos.

Roles relevantes

PythonIADataScienceBackendMachineLearningCyberSecurity

Severidad

Puntaje: 10.0(CRITICAL)
Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
AV: NETWORK
AC: LOW
PR: NONE
UI: NONE
S: CHANGED
C: HIGH
I: HIGH
A: HIGH
Tipo de falla (CWE): CWE-94

EPSS

Sin puntaje EPSS aún (CVE muy reciente).

Descripción técnica

MindsDB Minds Platform version 26.1.0 and earlier contains an unauthenticated remote code execution vulnerability that allows unauthenticated attackers to execute arbitrary OS commands by submitting crafted prompts to the unprotected POST /api/v1/responses/ endpoint, which reaches the Anton agent's scratchpad tool that calls exec() on attacker-influenced Python source without sandboxing. Attackers can first configure their own LLM API key through the unauthenticated PUT /api/v1/settings/ endpoint, then POST a prompt directing the agent to invoke the scratchpad tool with arbitrary Python code, achieving full OS command execution as the user running the desktop application and enabling access to SSH keys, stored credentials, and environment secrets.

Publicada: 14/8/2026, 19:18:01
Última modificación: 14/8/2026, 19:18:01

Referencias

InicioEventosBlogRecursosCursosEquipo