A critical authorization bypass vulnerability in OpenClaw allows authenticated attackers to execute arbitrary commands, posing a severe risk to developer workstations and CI infrastructure.

The gateway component fails to sanitize internal approval fields within node.invoke parameters. This allows any authenticated client to bypass system.run command execution gating and perform unauthorized operations on connected node hosts.

The vulnerability carries a CVSS score of 8.8, reflecting its high potential for system compromise. Successful exploitation grants an attacker the ability to execute arbitrary commands with the privileges of the underlying host, which could lead to complete system takeover, theft of intellectual property from developer workstations, or the compromise of CI/CD pipelines.

Immediate Action: Update the OpenClaw package to version 2026.2.14 or later to apply the necessary input sanitization fixes.

Proactive Monitoring: Review access logs for unusual node.invoke activity, specifically looking for anomalous parameters or unauthorized attempts to trigger system.run commands.

Compensating Controls: Implement strict network segmentation for node hosts to limit the blast radius of a potential compromise, and ensure that gateway credentials are managed using the principle of least privilege.

Public Exploit Available: Yes, a public proof-of-concept exists via the Orioning/CVE-2026-28466 GitHub repository.

Given the potential for remote code execution and the existence of a public proof-of-concept, organizations using OpenClaw should prioritize patching immediately. Failure to update to version 2026.2.14 leaves infrastructure vulnerable to command injection by any user with valid gateway access.