<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:webfeeds="http://webfeeds.org/rss/1.0"><channel><title>MLflow (3.10.0-3.16.x) - CraftedSignal Threat Feed</title><link>https://feed.craftedsignal.io/products/mlflow-3.10.0-3.16.x/</link><description>Trending threats, MITRE ATT&amp;CK coverage, and detection metadata. Fed continuously.</description><generator>Hugo</generator><language>en</language><managingEditor>hello@craftedsignal.io</managingEditor><webMaster>hello@craftedsignal.io</webMaster><lastBuildDate>Tue, 27 Oct 2026 22:07:17 +0000</lastBuildDate><atom:link href="https://feed.craftedsignal.io/products/mlflow-3.10.0-3.16.x/feed.xml" rel="self" type="application/rss+xml"/><image><url>https://feed.craftedsignal.io/favicon-32x32.png</url><title>CraftedSignal Threat Feed</title><link>https://feed.craftedsignal.io/</link><width>32</width><height>32</height></image><webfeeds:icon>https://feed.craftedsignal.io/favicon.svg</webfeeds:icon><item><title>MLflow: Arbitrary file read via model version path validation (CVE-2026-116690)</title><link>https://feed.craftedsignal.io/briefs/2026-10-mlflow-traversal/</link><pubDate>Tue, 27 Oct 2026 22:07:17 +0000</pubDate><author>hello@craftedsignal.io</author><guid isPermaLink="true">https://feed.craftedsignal.io/briefs/2026-10-mlflow-traversal/</guid><description>MLflow versions 3.10.0 through 3.16.x contain a path traversal vulnerability allowing remote attackers to read arbitrary files from the server filesystem via the model-versions/get-artifact API.</description><content:encoded><![CDATA[<p>MLflow versions 3.10.0 through 3.16.x contain a path traversal vulnerability in the <code>_create_model_version()</code> function. The application fails to properly validate source paths that lack a URL scheme, allowing remote users to supply server-local paths. An attacker can leverage the <code>/api/2.0/mlflow/model-versions/get-artifact</code> endpoint to access any file readable by the MLflow server process. This vulnerability impacts both unauthenticated instances and deployments utilizing basic authentication where the attacker possesses model registration privileges. This issue is resolved in MLflow version 3.17.0.</p>
<h2 id="impact">Impact</h2>
<p>Successful exploitation allows for unauthorized reading of sensitive files located on the MLflow server's filesystem, potentially exposing credentials, configuration data, or other sensitive system artifacts.</p>
<h2 id="recommendation">Recommendation</h2>
<ul>
<li>Upgrade MLflow to version 3.17.0 or later immediately.</li>
<li>Implement strict network access controls to limit access to the MLflow API to trusted IP addresses only.</li>
<li>Review MLflow server process permissions to ensure the service runs with the least privilege necessary, minimizing the impact of potential filesystem traversal.</li>
</ul>
]]></content:encoded><category domain="severity">high</category><category domain="type">advisory</category></item></channel></rss>