<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Security on Marc Dougherty</title><link>https://www.marcdougherty.com/tags/security/</link><description>Recent content in Security on Marc Dougherty</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><copyright>© 2026 Marc Dougherty</copyright><lastBuildDate>Wed, 20 May 2026 17:03:11 -0700</lastBuildDate><atom:link href="https://www.marcdougherty.com/tags/security/index.xml" rel="self" type="application/rss+xml"/><item><title>MCP Servers and OAuth Credentials</title><link>https://www.marcdougherty.com/2026/mcp-servers-and-oauth-credentials/</link><pubDate>Wed, 20 May 2026 17:03:11 -0700</pubDate><guid>https://www.marcdougherty.com/2026/mcp-servers-and-oauth-credentials/</guid><description>&lt;h2 class="relative group"&gt;Intro
&lt;div id="intro" class="anchor"&gt;&lt;/div&gt;
&lt;span
class="absolute top-0 w-6 transition-opacity opacity-0 -start-6 not-prose group-hover:opacity-100 select-none"&gt;
&lt;a class="text-primary-300 dark:text-neutral-700 !no-underline" href="#intro" aria-label="Anchor"&gt;#&lt;/a&gt;
&lt;/span&gt;
&lt;/h2&gt;
&lt;p&gt;In the previous article in this series, we used an MCP server to store and
version frequently used prompts. Building on this, our next step is to
add MCP Tools to perform work on our behalf. Tech news is full
of stories where well-intentioned engineers gave credentials to AI agents,
which then leaked keys or destroyed work. How can we enable our agents while
managing the risk?&lt;/p&gt;</description></item></channel></rss>