<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Tool Design on Andreas Nissen</title><link>https://andreasnissen.dev/tags/tool-design/</link><description>Recent content in Tool Design on Andreas Nissen</description><generator>Hugo</generator><language>en-us</language><copyright>© Andreas Nissen</copyright><lastBuildDate>Sat, 29 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://andreasnissen.dev/tags/tool-design/index.xml" rel="self" type="application/rss+xml"/><item><title>From Screenshots to Governed Tools</title><link>https://andreasnissen.dev/writing/from-screenshots-to-governed-tools/</link><pubDate>Sat, 29 Aug 2026 00:00:00 +0000</pubDate><guid>https://andreasnissen.dev/writing/from-screenshots-to-governed-tools/</guid><description>&lt;p&gt;An agent can use a website by looking at pixels, clicking controls, and reading the result. That is powerful because it works without changing the application.&lt;/p&gt;&#10;&lt;p&gt;It is also the least explicit way to automate a consequential interface.&lt;/p&gt;&#10;&lt;p&gt;The agent has to infer what a chart means, which control maps to which action, whether the page changed, and whether an approval actually exists. The human sees a visual interface. The agent reconstructs a hidden API from appearance and behavior.&lt;/p&gt;</description></item></channel></rss>