<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Mcp-Security on CuraDevOps</title><link>https://curadevops.metacog.co.kr/tags/mcp-security/</link><description>Recent content in Mcp-Security on CuraDevOps</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 18 Aug 2026 11:17:30 +0000</lastBuildDate><atom:link href="https://curadevops.metacog.co.kr/tags/mcp-security/index.xml" rel="self" type="application/rss+xml"/><item><title>Critical RCE in Serena MCP coding agent — update to 1.7.0</title><link>https://curadevops.metacog.co.kr/insights/2026-08-18-critical-remote-code-execution-in-serena-a-popular-mcp-codin/</link><pubDate>Tue, 18 Aug 2026 11:17:30 +0000</pubDate><guid>https://curadevops.metacog.co.kr/insights/2026-08-18-critical-remote-code-execution-in-serena-a-popular-mcp-codin/</guid><description>&lt;ul>
&lt;li>&lt;strong>Platform/SRE — Skip&lt;/strong>&lt;/li>
&lt;li>&lt;strong>CI/CD — Act:&lt;/strong> Published GHSA-pp25-4cg4-qcr9 details a critical server-side template injection in serena-agent ≤1.6.1 that executes arbitrary code via a malicious .serena/project.yml smuggled in any cloned repo — a direct supply-chain threat to developer and CI environments; upgrade to serena-agent 1.7.0 now.&lt;/li>
&lt;li>&lt;strong>Leader — Plan:&lt;/strong> This is an early, documented example of a new risk class: MCP servers embedded in the SDL grant LLMs broad filesystem and shell access, making any compromise severe; evaluate whether your AI coding-agent adoption policies explicitly address this attack surface before broader org rollout.&lt;/li>
&lt;/ul></description></item></channel></rss>