<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>CodeQuality on Time Friend</title><link>https://time-friend.com/en/tags/codequality/</link><description>Recent content in CodeQuality on Time Friend</description><generator>Hugo</generator><language>en-GB</language><lastBuildDate>Wed, 29 Jul 2026 15:15:07 +0800</lastBuildDate><atom:link href="https://time-friend.com/en/tags/codequality/index.xml" rel="self" type="application/rss+xml"/><item><title>Would You Merge Code an Agent Wrote Alone</title><link>https://time-friend.com/en/archive/claude-code-self-verification/</link><pubDate>Thu, 05 Feb 2026 09:45:00 +0000</pubDate><guid>https://time-friend.com/en/archive/claude-code-self-verification/</guid><description>&lt;p&gt;Claude finishes editing and says &amp;ldquo;Done.&amp;rdquo; You glance at it, looks fine, merge it. Then Monday comes and production is down.&lt;/p&gt;
&lt;p&gt;Without a runnable check, the agent&amp;rsquo;s &amp;ldquo;done&amp;rdquo; and your &amp;ldquo;done&amp;rdquo; are two different things. The most expensive bugs aren&amp;rsquo;t crashes — they&amp;rsquo;re silent wrong answers.&lt;/p&gt;
&lt;h2 id="the-simplest-check-one-line-in-your-prompt"&gt;The simplest check: one line in your prompt&lt;/h2&gt;
&lt;p&gt;Add this at the end of your prompt:&lt;/p&gt;
&lt;pre tabindex="0"&gt;&lt;code&gt;Run the tests after you finish. Keep fixing until they all pass.
&lt;/code&gt;&lt;/pre&gt;&lt;p&gt;Claude will run &lt;code&gt;npm test&lt;/code&gt; on its own, check output, fix what&amp;rsquo;s broken, and rerun. No input needed from you.&lt;/p&gt;</description></item></channel></rss>