CVEvulnerability
CVE-2026-2256
6.5 MEDIUM
Official page ↗
- Status
- Awaiting Analysis
- Published
- 2026-03-02
- Last modified
- 2026-06-17
- Assigner
- [email protected]
- CVSS 3.1 (134c704f-9b21-4f2e-91b3-4a467353bcc0)
- 6.5 MEDIUM — CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
- Weaknesses
- CWE-77
Description
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.
Related
Weakness · CWE (1)
- CWE-77 Improper Neutralization of Special Elements used in a Command weakness
References
- https://github.com/Itamar-Yochpaz/CVE-2026-2256-PoC github.com
- https://github.com/modelscope/ms-agent github.com
- https://medium.com/@itamar.yochpaz/cve-2026-2256-from-ai-prompt-to-full-system-compromise-a4114c718326 medium.com
- https://www.hiddenlayer.com/research/indirect-prompt-injection-of-claude-computer-use www.hiddenlayer.com
- https://www.kb.cert.org/vuls/id/431821 www.kb.cert.org