OpenClaw AI Assistant Compromised via WhatsApp: Three Critical Vulnerabilities Allowed Credential Theft, Sandbox Escape, and Arbitrary Code Execution on Host
Security researchers have disclosed three serious vulnerabilities in OpenClaw that could allow attackers to steal credentials, escalate privileges, and execute arbitrary code on systems running the AI assistant. The flaws affected both home users and corporate deployments, highlighting risks associated with insufficient input validation and sandbox isolation in AI tooling.
Command Filtering Failures (CVSS 8.8)
Two of the vulnerabilities received a CVSS score of 8.8. They were caused by incomplete command filtering mechanisms that did not block all malicious inputs. As a result, an attacker could inject system commands into contexts where they should have been prohibited, enabling unauthorized actions ranging from data exfiltration to privilege escalation.
Sandbox Bypass via Directory Mounting (CVSS 8.4)
The third vulnerability, rated 8.4, allowed attackers to circumvent sandbox restrictions by mounting parent directories. Although OpenClaw blocked direct access to sensitive folders such as ~/.ssh, ~/.aws, and ~/.gnupg, it could still permit mounting of the entire /home directory. This effectively placed SSH keys, cloud tokens, and GPG secrets directly accessible to the attacker.
Further escalation was possible through the /var directory, which could grant access to the Docker socket and enable escape from the sandbox directly onto the host system. Researcher Chinmohan Nayak demonstrated that the attack could be initiated remotely via an external message sent through WhatsApp (owned by Meta, designated as extremist and banned in Russia) without requiring any prior foothold on the target machine.
Impact and Recommendations
In the worst-case scenario, an attacker could exfiltrate sensitive data, establish persistence, and execute arbitrary code on the victim’s machine. Developers addressed all three issues in OpenClaw 2026.6.6. Users are strongly advised to:
- Update to the latest version immediately
- Enable sandboxing for all secondary sessions
- Remove exec from the list of allowed tools
- Narrow the list of trusted communication channels to the minimum necessary
These measures significantly reduce the attack surface and help prevent similar exploitation attempts in the future.
Related articles
Security Week 2644: Active Exploitation of PHP2Shell Vulnerabilities Confirmed
PHP developers released patches for two critical vulnerabilities collectively known as PHP2Shell on October 20, affecting PHP 8.2 through 8.5 as well as older branches. The flaws allow unauthenticated remote code execution on affected servers, which power approximately 70 percent of websites. Despite expectations of immediate exploitation, confirmed attacks with working exploits only began on October 25. CrowdStrike noted technical difficulties in weaponizing the issues in real-world environments. Only one-fifth of top Russian websites had applied the fixes by October 25. Additional incidents included compromised eMudhra certificates, a critical GitLab flaw, and a 16-year-old Broadcom router vulnerability.
Positive Technologies Offers Up to 500,000 Rubles for Bypassing PT Application Firewall
Positive Technologies has expanded its Standoff Bug Bounty program to include rewards for bypassing its web application firewalls. Researchers can earn between 150,000 and 500,000 rubles for demonstrating practical bypasses against PT Application Firewall PRO and PT Application Firewall Cloud. The program targets the core protective mechanisms, including traffic filtering, attack detection rules, and the management layer used by administrators. Submissions must include reproducible attack scenarios that show how malicious requests can pass through undetected, rather than relying on scanner reports or theoretical findings. Reward amounts vary by severity and confirmed business impact, with critical issues in the PRO version paying up to 500,000 rubles. The company plans to use validated findings to strengthen detection rules and improve overall product resilience.
Tailscale SSH Vulnerability Allows Root Access via Unix Socket Redirection
A researcher identified flaws in Tailscale that let an attacker with existing SSH access on one node escalate to root privileges and move laterally across the network. Erica Windisch of Cyberdione Labs demonstrated the issues in lab conditions against Tailscale 1.104.2 and Headscale 0.29.5. The core problem stems from privileged operations performed by the tailscaled process when handling Unix socket redirection, granting an ordinary SSH session access to root-only resources. The researcher used the technique to interact with Docker and run elevated commands, and showed that broad access policies allow automated lateral movement in under a second. Additional issues include denial-of-service conditions and bypasses of certain restrictions. Recommendations include reviewing SSH policies and disabling unnecessary socket redirection, noting that Headscale 0.29.5 lacks an independent way to turn the feature off.
Critical CVE-2026-111497 Out-of-Bounds Write Flaw in Google Brotli Decoder Impacts Broad Range of Environments
A severe vulnerability has been identified in the Google-developed Brotli decoder used for decompressing data in web communications and other applications. The flaw, tracked as CVE-2026-111497, allows out-of-bounds writes when processing specially crafted compressed data, potentially enabling arbitrary code execution, information disclosure, or denial of service. It carries a CVSS v4.0 base score of 9.2 and is rated Critical. Google has released patched versions 1.2.1, 1.1.1, and 1.0.10 of the library. Node.js incorporated the fix in its updates, while Microsoft issued an out-of-band patch for .NET on October 25, 2026, to address the issue. The vulnerability affects any environment relying on the Brotli C implementation for data decompression.