Paper/Archive/CVE-2025-67511

CVE-2025-67511

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

EPSS 2%public exploit CVSS 9.7

critical.is priorityof 100

CVSS 9.7 (CNA)9.7 / 10 × 3029.1 Not in CISA KEV+300.0 EPSS percentile 80.9.809 × 2016.2 Exploit: SSVC poc+1010.0 SSVC exploitation: poc+100.0

Description · from the CVE record

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.

Affected

ProductAffectedFixed inSource

cai<= 0.5.9–CNA

Scores by source

GitHub_M (CNA)9.7CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

NVD9.6CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

CISA ADPSSVCExploitation: poc · Automatable: no · Technical impact: total

References

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