<?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>Cpe:2.3:a:coze:coze_studio:*:*:*:*:*:*:*:* - CraftedSignal Threat Feed</title><link>https://feed.craftedsignal.io/cpes/cpe2.3acozecoze_studio/</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, 20 Oct 2026 02:32:17 +0000</lastBuildDate><atom:link href="https://feed.craftedsignal.io/cpes/cpe2.3acozecoze_studio/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>Remote Code Execution in Coze Studio via Workflow Nodes</title><link>https://feed.craftedsignal.io/briefs/2026-10-coze-studio-rce/</link><pubDate>Tue, 20 Oct 2026 02:32:17 +0000</pubDate><author>hello@craftedsignal.io</author><guid isPermaLink="true">https://feed.craftedsignal.io/briefs/2026-10-coze-studio-rce/</guid><description>Coze Studio versions prior to 0.5.0 allow authenticated users to execute arbitrary commands by bypassing sandbox protections in Python workflow nodes.</description><content:encoded><![CDATA[<p>Coze Studio versions prior to 0.5.0 contain an insecure default configuration that fails to properly sandbox Python code execution within workflow nodes. An authenticated attacker can manipulate these workflow nodes to execute arbitrary code within a subprocess. By bypassing the existing regex-based import filters - specifically using the <code>__import__('subprocess')</code> technique - an attacker can gain command execution on the host environment. This vulnerability poses a significant risk to the integrity and confidentiality of the Coze Studio instance, as successful exploitation allows an attacker to exfiltrate sensitive environment variables, extract model API keys, and pivot to connected backend services such as MySQL, MinIO, Redis, and etcd.</p>
<h2 id="impact">Impact</h2>
<p>Successful exploitation of CVE-2026-77934 enables an attacker to achieve remote code execution as the user running the Coze Studio service. This access facilitates the theft of credentials (API keys) and unauthorized access to internal data stores (MySQL, MinIO, Redis, etcd), potentially resulting in full database compromise or cross-environment lateral movement.</p>
<h2 id="recommendation">Recommendation</h2>
<ol>
<li>Upgrade Coze Studio to version 0.5.0 or later to ensure proper sandboxing of Python code execution.</li>
<li>Review logs for unexpected <code>subprocess</code> or <code>os.system</code> calls originating from the service process.</li>
<li>Restrict network access for the Coze Studio backend infrastructure to prevent unauthorized access to internal databases (MySQL, Redis, etcd, MinIO) if the process is compromised.</li>
</ol>
]]></content:encoded><category domain="severity">high</category><category domain="type">advisory</category><category>remote-code-execution</category><category>sandbox-escape</category><category>python</category></item></channel></rss>