Machine-written summary
Summary written by Claude Haiku from the source advisory. Not reviewed by a person.
The ms-agent package, versions 1.6.0rc1 and earlier, is vulnerable to command injection. An attacker who can supply crafted input that reaches the agent's processing can run arbitrary operating system commands. The source gives few details about the required access or the exact input path. No fixed version is listed, so no upgrade target is available.
Not analysed yet
We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.
Severity and exploitation
| Score | Rating | CVSS | Vector | Source |
|---|---|---|---|---|
| 6.5 | Medium | 3.1 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N | GitHub Advisory Database |
- CISA KEV Not listed.
- EPSS 0.021, the 81st percentile, on 15 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.
Timeline
- 31 Oct 2026, 15:02 UTC First seen by agentprime
- 31 Oct 2026, 15:02 UTC Confirmed by OSV
Sources
- GitHub Advisory Database: GHSA-4gc2-344q-r2rw
- OSV: GHSA-4gc2-344q-r2rw
- nvd.nist.gov/vuln/detail/CVE-2026-2256
- github.com/Itamar-Yochpaz/CVE-2026-2256-PoC
- github.com/modelscope/ms-agent
- medium.com/@itamar.yochpaz/cve-2026-2256-from-ai-prompt-to-full-system-compromise-a4114c718326
- www.hiddenlayer.com/research/indirect-prompt-injection-of-claude-computer-use
- www.kb.cert.org/vuls/id/431821