CVE-2026-2256
Command injection vulnerability in ModelScope's ms-agent
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.
Attack profile
How easy it is to exploit
- Attack vector
- Network
- Complexity
- Low
- Privileges needed
- None
- User interaction
- None
What an attacker can affect
- Confidentiality
- Low
- Integrity
- Low
- Availability
- None
Exploit likelihood
What each source reports
CVE List Score used
6.5
Command injection vulnerability in ModelScope's ms-agent
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.
- Record
CVE-2026-2256- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Weakness
- CWE-77
- Updated
- 2026-11-02
EUVDEUVD-2026-9257
6.5
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.
- Record
EUVD-2026-9257- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Updated
- 2026-11-02
GitHub AdvisoriesGHSA-4gc2-344q-r2rw
6.5
MS-Agent vulnerable to Command Injection
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.
- Record
GHSA-4gc2-344q-r2rw- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Weakness
- CWE-77
- Updated
- 2026-11-02
NVD
6.5
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.
- Record
CVE-2026-2256- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Weakness
- CWE-77
- Updated
- 2026-11-02
OSV
6.5
Command injection vulnerability in ModelScope's ms-agent
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.
- Record
CVE-2026-2256- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Updated
- 2026-11-02
OSVGHSA-4gc2-344q-r2rw
6.5
MS-Agent vulnerable to Command Injection
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.
- Record
GHSA-4gc2-344q-r2rw- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Weakness
- CWE-77
- Updated
- 2026-11-02
OSVPYSEC-2026-2668
6.5
MS-Agent vulnerable to Command Injection
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.
- Record
PYSEC-2026-2668- CVSS 3.1
- 6.5 Medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N - Updated
- 2026-11-02
Affected products
Affected packages
Public exploits
Links only. Every CVE stores no exploit code.
| Source | Link | Date |
|---|---|---|
| PoC in GitHub | Itamar-Yochpaz/CVE-2026-2256-PoC | 2026-02-09 |
| PoC in GitHub | melbratic/CVE-2026-2256-Threat-Model----ms-agent-Command-Injection | 2026-06-03 |
| PoC in GitHub | mruniversity/CVE-2026-2256- | 2026-06-04 |
History
| Recorded | Changed | Sources |
|---|---|---|
| first indexed | CVE List, EUVD, GitHub Advisories, NVD, OSV, PoC in GitHub |
Dates are when Every CVE recorded each change. Upstream dates may differ.
In short
CVE-2026-2256 is a medium vulnerability (CVSS 6.5) in Modelscope Ms-agent, published March 2, 2026. Public exploit code exists. EPSS puts the chance of exploitation in the next 30 days at 1.70%, the 77th percentile. No source names a fixed version.
References
github.com/modelscope/ms-agent
hiddenlayer.com/research/indirect-prompt-injection-of-claude-computer-use
medium.com/@itamar.yochpaz/cve-2026-2256-from-ai-prompt-to-full-system-compromise-a4114c718326
github.com/Itamar-Yochpaz/CVE-2026-2256-PoC
kb.cert.org/vuls/id/431821
nvd.nist.gov/vuln/detail/CVE-2026-2256
github.com/advisories/GHSA-4gc2-344q-r2rw
github.com/CVEProject/cvelistV5/tree/main/cves/2026/2xxx/CVE-2026-2256.json
pypi.org/project/ms-agent