An improper trust boundary vulnerability in OpenClaw allows for unauthenticated remote code execution, posing a significant risk to workspace integrity.

This vulnerability, classified as CWE-829, involves an improper trust boundary that permits untrusted workspace channel shadows to execute code during built-in channel setup and login. An unauthenticated attacker can clone a workspace and inject a malicious plugin to achieve in-process code execution before the plugin is explicitly trusted.

The vulnerability carries a CVSS score of 7.8, indicating a high severity level. Successful exploitation grants an attacker the ability to execute arbitrary code within the context of the application, potentially leading to full compromise of the workspace environment, theft of sensitive data, or unauthorized lateral movement within the host system.

Immediate Action: Update the OpenClaw package to version 2026.4.2 or later to apply the necessary security fixes for the improper trust boundary.

Proactive Monitoring: Review application and workspace access logs for unusual plugin initialization patterns or unexpected channel setup activity.

Compensating Controls: Restrict the ability of users to import or clone external workspaces from untrusted sources until the software has been updated.

Public Exploit Available: No (exploit_available: unknown)

Given the potential for complete system compromise, administrators should prioritize updating OpenClaw to version 2026.4.2 immediately. Ensuring that all workspaces are managed within a trusted environment is essential to preventing exploitation while the patching process is completed.