<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>MCP on PG Blog</title><link>https://pg-blogs.netlify.app/tags/mcp/</link><description>Recent content in MCP on PG Blog</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sat, 04 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://pg-blogs.netlify.app/tags/mcp/index.xml" rel="self" type="application/rss+xml"/><item><title>The Model Context Protocol in Java</title><link>https://pg-blogs.netlify.app/posts/28-model-context-protocol-in-java/</link><pubDate>Sat, 04 Jul 2026 00:00:00 +0000</pubDate><guid>https://pg-blogs.netlify.app/posts/28-model-context-protocol-in-java/</guid><description>&lt;h2 id="introduction"&gt;Introduction&lt;/h2&gt;
&lt;p&gt;Every agent needs tools, and every tool needs a way to reach the model. &lt;a href="https://pg-blogs.netlify.app/posts/14-building-agentic-workflows-in-java/"&gt;Building Agentic Workflows in Java&lt;/a&gt; built that connection by hand — a hand-written &lt;code&gt;Tool&lt;/code&gt; schema, a loop that dispatches on &lt;code&gt;toolUse.name()&lt;/code&gt;. &lt;a href="https://pg-blogs.netlify.app/posts/26-llm-frameworks-vs-the-raw-sdk-in-java/"&gt;LLM Frameworks vs. the Raw SDK in Java&lt;/a&gt; showed LangChain4j and Spring AI turning an annotated Java method into that same schema via reflection. Both are still &lt;em&gt;bespoke&lt;/em&gt;: the tool lives inside one process, wired to one agent, in one language.&lt;/p&gt;</description></item><item><title>The Model Context Protocol in Python</title><link>https://pg-blogs.netlify.app/posts/29-model-context-protocol-in-python/</link><pubDate>Sat, 04 Jul 2026 00:00:00 +0000</pubDate><guid>https://pg-blogs.netlify.app/posts/29-model-context-protocol-in-python/</guid><description>&lt;h2 id="introduction"&gt;Introduction&lt;/h2&gt;
&lt;p&gt;Every agent needs tools, and every tool needs a way to reach the model. &lt;a href="https://pg-blogs.netlify.app/posts/15-building-agentic-workflows-in-python/"&gt;Building Agentic Workflows in Python&lt;/a&gt; built that connection by hand — a hand-written JSON schema, a loop that dispatches on &lt;code&gt;block.name&lt;/code&gt;. &lt;a href="https://pg-blogs.netlify.app/posts/27-llm-frameworks-vs-the-raw-sdk-in-python/"&gt;LLM Frameworks vs. the Raw SDK in Python&lt;/a&gt; showed LangChain&amp;rsquo;s &lt;code&gt;@tool&lt;/code&gt; turning a plain function into that same schema via &lt;code&gt;bind_tools&lt;/code&gt;. Both are still &lt;em&gt;bespoke&lt;/em&gt;: the tool lives inside one process, wired to one agent, in one language.&lt;/p&gt;</description></item></channel></rss>