This vulnerability exists in the hermes-agent product by NousResearch, specifically in the detect_dangerous_command function within tools/approval.py. The flaw allows an attacker to manipulate input to cause OS command injection remotely, potentially leading to arbitrary command execution on the affected system. The vulnerability is present in version 5157f5427f19488b31c6fdebbacd15d798ce7f63 and earlier. The vendor was contacted but did not respond, and no official fix or patch has been published. The CVSS 4.0 base score is 6.9, indicating a medium severity with network attack vector, no privileges or user interaction required, and low to limited impact on confidentiality, integrity, and availability.

Successful exploitation allows remote attackers to execute arbitrary operating system commands on the affected hermes-agent, potentially compromising system integrity and confidentiality. The impact is rated medium severity with limited impact on confidentiality, integrity, and availability. There are no known exploits in the wild at this time.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Since the vendor has not responded and no patch or workaround is available, users should consider restricting network access to the affected hermes-agent instances and monitor for suspicious activity related to command execution. Avoid exposing the vulnerable component to untrusted networks until a fix is available.