A critical allowlist bypass in OpenClaw enables remote authenticated operators to execute arbitrary PowerShell commands, circumventing existing security restrictions.

This is an allowlist bypass vulnerability in the handling of PowerShell encoded-commands. Attackers can leverage unrecognized flag aliases to bypass the parser's security checks and execute arbitrary PowerShell content.

The ability to execute arbitrary PowerShell commands provides an attacker with a powerful vehicle for system-wide compromise. With a CVSS score of 8.8, this flaw could lead to data theft, lateral movement, or complete system takeover, causing severe damage to business operations.

Immediate Action: Patch OpenClaw to version 2026.5.12 or higher to correct the PowerShell command parsing logic.

Proactive Monitoring: Monitor for suspicious PowerShell execution patterns or the use of unusual encoded command syntax in logs.

Compensating Controls: Utilize endpoint detection and response (EDR) solutions to flag or block unauthorized PowerShell scripts and encoded commands.

Public Exploit Available: false

Given that this vulnerability allows for the execution of arbitrary commands, it presents a significant risk to the integrity of the host system. Immediate patching to version 2026.5.12 is required to mitigate this bypass and prevent unauthorized command execution.