CVE-2025-67511

Cybersecurity AI (CAI) vulnerable to Command Injection in run_ssh_command_with_credentials Agent tool

CVSS 9.7 CRITICALEPSS 1.8%CWE-77

Vexday Risk Score

48Attention

SSVC decision (CISA)

Attend

PoC available → attend closely

CVSS 9.7EPSS 1.8%KEV nãoPoC públicaNuclei —Metasploit —Patch —

Lifecycle

09 Dec 2025Public PoC

10 Dec 2025Published on NVD

Recommendation: Plan a near-term fix — a public PoC already exists.

Cybersecurity AI (CAI) is an open-source framework for building and deploying AI-powered offensive and defensive automation. Versions 0.5.9 and below are vulnerable to Command Injection through the run_ssh_command_with_credentials() function, which is available to AI agents. Only password and command inputs are escaped in run_ssh_command_with_credentials to prevent shell injection; while username, host and port values are injectable. This issue does not have a fix at the time of publication.

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

Affected products

aliasrobotics · cai

public PoCs found — 1

githubgithub.com/edoardottt/CVE-2025-67511★ 4

⚠ Public resources, to assess the exposure of systems you control or are authorized to test. Test only with authorization.

References

https://github.com/aliasrobotics/cai/commit/09ccb6e0baccf56c40e6cb429c698750843a999c https://github.com/aliasrobotics/cai/security/advisories/GHSA-4c65-9gqf-4w8h https://www.hacktivesecurity.com/blog/2025/12/10/cve-2025-67511-tricking-a-security-ai-agent-into-pwning-itself