<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Openadapt-Flow on OpenAdapt Blog</title><link>https://blog.openadapt.ai/tags/openadapt-flow/</link><description>Recent content in Openadapt-Flow on OpenAdapt Blog</description><generator>Hugo -- 0.164.0</generator><language>en-us</language><lastBuildDate>Wed, 08 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.openadapt.ai/tags/openadapt-flow/index.xml" rel="self" type="application/rss+xml"/><item><title>The 500th run: compiled automation vs. computer-use agents</title><link>https://blog.openadapt.ai/posts/the-500th-run/</link><pubDate>Wed, 08 Jul 2026 00:00:00 +0000</pubDate><guid>https://blog.openadapt.ai/posts/the-500th-run/</guid><description>Same task, same success check: 100/100 for the compiled script, 20/20 for the agent. The difference was 4.9 s vs 37.5 s per run, and $0 vs $0.27.</description></item></channel></rss>