<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
    <channel>
        <title>Agent - Tag - Simi Studio</title>
        <link>/en/tags/agent/</link>
        <description>Agent - Tag - Simi Studio</description>
        <generator>Hugo -- gohugo.io</generator><language>en</language><managingEditor>simi@simi.studio (Simi)</managingEditor>
            <webMaster>simi@simi.studio (Simi)</webMaster><lastBuildDate>Tue, 14 Apr 2026 10:00:00 &#43;0800</lastBuildDate><atom:link href="/en/tags/agent/" rel="self" type="application/rss+xml" /><item>
    <title>When AI Agents Go Off the Rails: From Model Worship to Environment Engineering</title>
    <link>/en/posts/harness-engineering/</link>
    <pubDate>Tue, 14 Apr 2026 10:00:00 &#43;0800</pubDate>
    <author>simi@simi.studio (Simi)</author>
    <guid>/en/posts/harness-engineering/</guid>
    <description><![CDATA[Models keep getting stronger, yet AI Agents still 'forget,' go out of bounds, and repeat mistakes in production. The problem isn't the model—it's how little engineering effort goes into the runtime environment. Harness Engineering, emerging in 2026, tries to fix this.]]></description>
</item>
<item>
    <title>2025 AI Coding: Tool Landscape Settles, Engineering Practices Mature</title>
    <link>/en/posts/ai-coding-2025-state-of-the-art/</link>
    <pubDate>Mon, 15 Dec 2025 10:00:00 &#43;0800</pubDate>
    <author>simi@simi.studio (Simi)</author>
    <guid>/en/posts/ai-coding-2025-state-of-the-art/</guid>
    <description><![CDATA[2025 was the year AI coding went from chaos to order. Tool landscape settled, engineering practices accumulated, evaluation standards established. This article records my observations on AI coding's evolution this year.]]></description>
</item>
<item>
    <title>MCP in Practice: Connecting AI Agents to Your Toolchain</title>
    <link>/en/posts/mcp-model-context-protocol/</link>
    <pubDate>Thu, 10 Apr 2025 10:00:00 &#43;0800</pubDate>
    <author>simi@simi.studio (Simi)</author>
    <guid>/en/posts/mcp-model-context-protocol/</guid>
    <description><![CDATA[MCP (Model Context Protocol) is Anthropic's open protocol for connecting AI Agents to external tools in a standardized way. This article explains what MCP is, how to use it, and its real value.]]></description>
</item>
<item>
    <title>Spec First, Code Second: Using OpenSpec to Constraint Your AI Coding Assistant</title>
    <link>/en/posts/openspec-spec-driven-ai-development/</link>
    <pubDate>Sat, 15 Mar 2025 10:00:00 &#43;0800</pubDate>
    <author>simi@simi.studio (Simi)</author>
    <guid>/en/posts/openspec-spec-driven-ai-development/</guid>
    <description><![CDATA[OpenSpec is a Spec-Driven Development framework that solves the problem of vague requirements and unpredictable AI output by establishing a spec layer before AI writes any code.]]></description>
</item>
</channel>
</rss>
