<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agent Engineering Notes on E7Coding</title><link>https://www.e7coding.com/en/series/agent-engineering-notes/</link><description>Recent content in Agent Engineering Notes on E7Coding</description><generator>Hugo</generator><language>en</language><managingEditor>Joy</managingEditor><webMaster>Joy</webMaster><lastBuildDate>Sat, 11 Jul 2026 10:30:00 +0800</lastBuildDate><atom:link href="https://www.e7coding.com/en/series/agent-engineering-notes/index.xml" rel="self" type="application/rss+xml"/><item><title>Agent Skills: Packaging Senior Engineering Discipline Into AI Coding Agents</title><link>https://www.e7coding.com/en/posts/addyosmani-agent-skills-engineering-discipline/</link><pubDate>Sat, 11 Jul 2026 10:30:00 +0800</pubDate><author>Joy</author><guid>https://www.e7coding.com/en/posts/addyosmani-agent-skills-engineering-discipline/</guid><description>&lt;blockquote&gt;
&lt;p&gt;AI coding agents have a common failure mode: they default to the &lt;strong&gt;shortest path&lt;/strong&gt;. Ask one to add a feature, and it often starts coding immediately, skipping specs, tests, security review, and the practices that make software reliable. &lt;a href="https://github.com/addyosmani" target="_blank" rel="noopener noreferrer"&gt;Addy Osmani&lt;/a&gt;&amp;rsquo;s &lt;strong&gt;&lt;a href="https://github.com/addyosmani/agent-skills" target="_blank" rel="noopener noreferrer"&gt;Agent Skills&lt;/a&gt;&lt;/strong&gt; aims to fix exactly that: package senior engineers&amp;rsquo; production workflows, quality gates, and judgment into &lt;strong&gt;skills that agents can follow consistently&lt;/strong&gt;, covering every stage from idea to launch.&lt;/p&gt;</description></item><item><title>What Is A2A: How Different Agents Discover, Communicate, and Collaborate</title><link>https://www.e7coding.com/en/posts/agent2agent-a2a-protocol-introduction/</link><pubDate>Wed, 01 Jul 2026 15:40:00 +0800</pubDate><author>Joy</author><guid>https://www.e7coding.com/en/posts/agent2agent-a2a-protocol-introduction/</guid><description>&lt;p&gt;A2A stands for &lt;strong&gt;Agent2Agent Protocol&lt;/strong&gt;.&lt;/p&gt;
&lt;p&gt;It is not about how a model calls tools. It is about how different agents discover each other, exchange messages, and collaborate on tasks.&lt;/p&gt;
&lt;p&gt;As organizations begin deploying multiple agents internally, a practical problem quickly appears: a sales agent, support agent, finance agent, legal agent, and engineering agent may come from different teams, vendors, and frameworks. They cannot only operate inside their own systems. They also need to delegate work, pass context, and return results to one another.&lt;/p&gt;</description></item><item><title>What Is MCP: A Stable Way for Agents to Connect Tools and Context</title><link>https://www.e7coding.com/en/posts/model-context-protocol-mcp-introduction/</link><pubDate>Wed, 01 Jul 2026 15:30:00 +0800</pubDate><author>Joy</author><guid>https://www.e7coding.com/en/posts/model-context-protocol-mcp-introduction/</guid><description>&lt;p&gt;MCP stands for &lt;strong&gt;Model Context Protocol&lt;/strong&gt;.&lt;/p&gt;
&lt;p&gt;Its role is clear: it provides an open protocol for connecting AI applications to external systems. Those systems may be local files, databases, search engines, business APIs, design tools, knowledge bases, or predefined prompts and workflows.&lt;/p&gt;</description></item></channel></rss>