Network Policy Bypass in OpenClaw Sandbox Exec-Server

Executive Summary

OpenClaw versions before 2026.6.6 have a network policy bypass vulnerability in the sandbox exec-server. This allows lower-trust callers to send HTTP requests through the exec-server to access internal network destinations that should have been blocked by OpenClaw's configured policies.

Useful 0 Not Useful 0

Detection Guidance

To detect this vulnerability, monitor network traffic for unauthorized HTTP requests originating from the OpenClaw exec-server. Check logs for requests to internal network destinations that should be blocked by policy. Verify if lower-trust callers can access the exec-server feature.

Useful 0 Not Useful 0

Impact Analysis

Attackers could exploit this to reach restricted internal network resources, potentially accessing sensitive data or systems. The impact depends on configuration and whether lower-trust input can access the affected exec-server path. The vulnerability is classified as SSRF with a CVSS score of 7.7 (High).

Useful 0 Not Useful 0

Compliance Impact

This vulnerability could potentially violate compliance with GDPR and HIPAA by allowing unauthorized access to internal network resources. GDPR requires strict data access controls and breach notifications, while HIPAA mandates safeguards for protected health information. A network policy bypass could lead to unauthorized data exposure, triggering compliance violations if sensitive data is accessed or exfiltrated.

Useful 0 Not Useful 0

Mitigation Strategies

Upgrade OpenClaw to version 2026.6.6 or later. Restrict exec-server access to trusted operators only. Disable the exec-server feature if not required. Implement stricter allowlists and avoid shared Gateways for untrusted users.

Useful 0 Not Useful 0