<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Evidence on arclab</title><link>https://blog.arclab.dev/tags/evidence/</link><description>Recent content in Evidence on arclab</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sat, 29 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.arclab.dev/tags/evidence/index.xml" rel="self" type="application/rss+xml"/><item><title>My coding agents run unsupervised. Their claims don't.</title><link>https://blog.arclab.dev/posts/coding-agents-run-unsupervised/</link><pubDate>Sat, 29 Aug 2026 00:00:00 +0000</pubDate><guid>https://blog.arclab.dev/posts/coding-agents-run-unsupervised/</guid><description>&lt;p&gt;One morning in April 2026, I was reviewing a pull request that one of my coding agents had put together overnight. The report was genuinely good: what changed and why, which tests covered it, one edge case it had deliberately punted on, with a note explaining the punt. It read like the work of a careful colleague. I was about to hit merge when it occurred to me what I was actually doing: accepting a story about the work, written by the thing that did the work, as if the story were the work.&lt;/p&gt;</description></item></channel></rss>