This vulnerability in AgentDeskAI browser-tools-mcp (versions 1.0 through 1.2.0) allows remote attackers to perform OS command injection via improper handling in browser-tools-server/browser-connector.ts. The issue enables execution of arbitrary operating system commands due to insufficient input validation or sanitization. The vulnerability is publicly known and exploitable remotely without privileges or user interaction. No official fix or remediation guidance has been provided by the vendor as of the published date.

Successful exploitation could allow an attacker to execute arbitrary OS commands on the affected system remotely, potentially leading to system compromise or unauthorized control. The CVSS score of 6.9 indicates a medium severity impact with low attack complexity and no required privileges or user interaction. However, the vendor has not yet issued a patch or mitigation, increasing exposure risk.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Since the vendor has not responded or released a fix, users should consider applying workarounds such as restricting network access to the affected component or employing runtime application controls to detect and block suspicious command execution attempts until an official patch is available.