A critical command injection vulnerability in Google OpenClaw allows authenticated attackers to execute arbitrary code on paired nodes, potentially leading to full system compromise.

The software fails to enforce execution approval paths for the googlemeet.chrome command, which permits a low-privileged authenticated agent to supply malicious command arrays that execute directly on a paired node.

A successful exploit grants an attacker the ability to execute arbitrary code on the target node, resulting in unauthorized access to sensitive files, credentials, and browser profiles. Given the CVSS score of 8.8, the risk is classified as High, as it permits full loss of confidentiality, integrity, and availability for the affected node. This could lead to significant data breaches and lateral movement within the network.

Immediate Action: Update OpenClaw to version 2026.7.1 or later to apply the necessary command execution approval logic.

Proactive Monitoring: Review system logs for unusual process execution patterns or unexpected calls to the googlemeet.chrome command originating from automated agents.

Compensating Controls: If immediate patching is not feasible, administrators should disable the Google Meet plugin or remove the googlemeet.chrome command from the list of allowed node commands to prevent exploitation.

Public Exploit Available: No

This vulnerability represents a significant security risk due to the potential for remote code execution on paired infrastructure. Organizations should prioritize the deployment of version 2026.7.1 across all affected environments to eliminate the underlying command injection vector. If updates cannot be applied immediately, the suggested workaround of disabling the Google Meet plugin is highly recommended to secure the environment.