CVE-2026-119328
Reporte público de vulnerabilidad
BAJA Exploit público detectado CVSS 6.5 / MEDIUM CWE-22GTN Risk Score
Monitoreo
Score calculado con CVSS, EPSS, CISA KEV, evidencia pública de exploit y vector de ataque.
Resumen ejecutivo
Resumen ejecutivo
Immich is a high-performance self-hosted photo and video management solution. Prior to 3.3.0, the unauthenticated machine-learning /predict endpoint in machine-learning/immich_ml/main.py accepted an options object and passed its values into model construction in machine-learning/immich_ml/models/base.py without restricting security-sensitive parameters. An attacker with network access to the service could set cache_dir to an arbitrary writable path, causing model downloads to write there and causing clear_cache() to pass that path to shutil.rmtree() after selected model-loading failures. Depending on filesystem permissions and the selected path, this could modify files in mounted data locations or recursively delete an attacker-selected directory. The model-name allowlist prevents this issue from becoming arbitrary-model deserialization or code execution. This issue is fixed in version 3.3.0. Resumen de riesgo: CVE-2026-119328 presenta CVSS 6.5, severidad MEDIA, EPSS 0.44%, CISA KEV: No, exploit público detectado: Sí. Producto/proveedor asociado: No disponible / No disponible. Se recomienda validar exposición, revisar versiones afectadas, aplicar parche o mitigación disponible y monitorear actividad relacionada.
Indicadores
¿De qué se trata?
Immich is a high-performance self-hosted photo and video management solution. Prior to 3.3.0, the unauthenticated machine-learning /predict endpoint in machine-learning/immich_ml/main.py accepted an options object and passed its values into model construction in machine-learning/immich_ml/models/base.py without restricting security-sensitive parameters. An attacker with network access to the service could set cache_dir to an arbitrary writable path, causing model downloads to write there and causing clear_cache() to pass that path to shutil.rmtree() after selected model-loading failures. Depending on filesystem permissions and the selected path, this could modify files in mounted data locations or recursively delete an attacker-selected directory. The model-name allowlist prevents this issue from becoming arbitrary-model deserialization or code execution. This issue is fixed in version 3.3.0.
Producto / proveedor
Fechas
Estado de explotación
Evidencia encontrada
| Fuente | Estado | Detalle | Confianza |
|---|---|---|---|
| NVD References | Referencia sospechosa/PoC | NVD contiene una referencia relacionada con PoC, exploit o repositorio público. | Media |
| NVD References | Referencia sospechosa/PoC | NVD contiene una referencia relacionada con PoC, exploit o repositorio público. | Media |
| NVD References | Referencia sospechosa/PoC | NVD contiene una referencia relacionada con PoC, exploit o repositorio público. | Media |
| NVD References | Referencia sospechosa/PoC | NVD contiene una referencia relacionada con PoC, exploit o repositorio público. | Media |
Nota: esta sección es referencial y defensiva. La ausencia de evidencia pública no garantiza que no exista explotación privada o no publicada.
¿Cómo superarlo / mitigarlo?
- Mantener monitoreo. La probabilidad actual de explotación es baja según EPSS.
- Vector de red: revisar exposición pública, reglas de firewall, WAF/VPN y segmentación.