← back
CVE-2026-2256

Command injection vulnerability in ModelScope's ms-agent

CVSS 6.5 MEDIUMEPSS 1.6%CWE-77
A command injection vulnerability in ModelScope's ms-agent versions v1.6.0rc1 and earlier exists, allowing an attacker to execute arbitrary operating system commands through crafted prompt-derived input.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
Affected products
ModelScope · ms-agent

Want to know if your infrastructure is exposed to this?

Talk to TrueHacking →
References
https://github.com/Itamar-Yochpaz/CVE-2026-2256-PoChttps://github.com/modelscope/ms-agenthttps://medium.com/@itamar.yochpaz/cve-2026-2256-from-ai-prompt-to-full-system-compromise-a4114c718326https://www.hiddenlayer.com/research/indirect-prompt-injection-of-claude-computer-usehttps://www.kb.cert.org/vuls/id/431821