Attack Chain

1. The attacker injects malicious JavaScript code into the SandboxJS environment.
2. The injected code uses Function.caller to obtain a reference to the internal LispType.Call runtime callback.
3. The attacker crafts a fake context and obj to pass as arguments to the leaked callback.
4. The crafted context includes a capture function that intercepts the internal function call within SandboxJS.
5. The attacker uses the captured function to leak static properties of the Object constructor.
6. The attacker obtains a reference to the host Function constructor by calling internal primitive functions.
7. The attacker crafts a string containing JavaScript code to execute on the host system.
8. The attacker invokes the host Function constructor with the malicious JavaScript code, resulting in remote code execution on the host.

Impact

Successful exploitation of this vulnerability (CVE-2026-43898) allows an attacker to bypass the SandboxJS sandbox and execute arbitrary JavaScript code on the host system. This can lead to complete system compromise, including data theft, malware installation, and denial-of-service. Given the nature of sandboxes being used to execute untrusted code, the impact is typically critical. The vulnerability affects all users of @nyariv/sandboxjs versions 0.9.5 and earlier.

Recommendation

• Upgrade the @nyariv/sandboxjs package to a version greater than 0.9.5 to patch CVE-2026-43898.
• Deploy the Sigma rule “Detect SandboxJS Function Caller Abuse” to detect attempts to exploit the Function.caller vulnerability.
• Monitor JavaScript execution within SandboxJS environments for unexpected calls to Function.caller.