<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ecosystem on Start AI Tools - Presented by Intent Solutions</title><link>https://startaitools.com/tags/ecosystem/</link><description>Recent content in Ecosystem on Start AI Tools - Presented by Intent Solutions</description><generator>Hugo</generator><language>en-US</language><copyright>Intent Solutions. All rights reserved.</copyright><lastBuildDate>Thu, 04 Jun 2026 22:27:52 -0600</lastBuildDate><atom:link href="https://startaitools.com/tags/ecosystem/index.xml" rel="self" type="application/rss+xml"/><item><title>From one adopter to two: the discovery-affordance spec just got named</title><link>https://startaitools.com/posts/discovery-affordance-second-adopter/</link><pubDate>Thu, 04 Jun 2026 08:00:00 -0500</pubDate><guid>https://startaitools.com/posts/discovery-affordance-second-adopter/</guid><description>&lt;p&gt;A few weeks ago I shipped schema 3.4.0 on &lt;a href="https://github.com/jeremylongshore/claude-code-plugins-plus-skills"&gt;claude-code-plugins-plus-skills&lt;/a&gt;. The headline feature: a single machine-crawlable manifest indexing 2,783 skills, ~97 KB gzipped. Any client — a marketplace UI, a CLI installer, a federated search index — can browse or search the catalog without scraping the repo directory by directory.&lt;/p&gt;
&lt;p&gt;The point of the manifest wasn&amp;rsquo;t &amp;ldquo;here is a better README.&amp;rdquo; It was a bet that skill ecosystems past a few hundred entries need an affordance for clients, not just humans. Without one, every new tool has to write a custom crawler against each repo&amp;rsquo;s layout. With one, the work happens once at the spec layer and any compliant index is queryable.&lt;/p&gt;</description></item></channel></rss>