Dask Distributed Remote Code Execution Vulnerability via Jupyter Proxy

A remote code execution vulnerability has been identified in Dask Distributed versions prior to 2026.1.0. This issue arises when Jupyter Lab, jupyter-server-proxy, and Dask Distributed are used together. Attackers can create a URL that exploits a cross-site scripting (XSS) vulnerability in the Dask dashboard, leading to code execution in Jupyter's default Python kernel. The attack requires Jupyter Lab and Dask to be running on localhost with default ports. When the crafted link is clicked, it triggers the execution of malicious code through the Jupyter Lab proxy into the Dask Dashboard.

Impact

Exploitation of this vulnerability allows for remote code execution on the server where Jupyter is running, executed in the context of the user.

Reproduction

To reproduce this vulnerability, first ensure that Jupyter Lab is running on the default port, along with jupyter-server-proxy, and that Dask Distributed is active on its default port. Then, click on a phishing URL crafted to exploit the XSS vulnerability in the Dask dashboard. This will open an error page through the Jupyter Lab proxy, triggering the execution of the malicious code in the Jupyter Python kernel.

Remediation

Users should upgrade to Dask Distributed version 2026.1.1 or later. Instructions for upgrading can be found in the Dask documentation.