<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:media="http://search.yahoo.com/mrss/"><channel>
<title>AgentRags: Cline</title><link>https://agentrags.com/signal/</link><description>Nightly picks that mention Cline.</description>
<atom:link href="https://agentrags.com/feeds/cline/signal.xml" rel="self" type="application/rss+xml"/><language>en</language>
<lastBuildDate>Sun, 08 Nov 2026 00:19:58 -0500</lastBuildDate><ttl>180</ttl>
<item><title>Hermes OpenAI-Compatible API + Reasoning Streaming Makes Other Apps Doorways</title><link>https://agentrags.com/story/8b58a8a6bb/</link><guid isPermaLink="true">https://agentrags.com/story/8b58a8a6bb/</guid><pubDate>Tue, 22 Sep 2026 23:00:00 -0400</pubDate><category>Tools &amp; Plugins</category><description><![CDATA[<p>Hermes can expose any configured agent (skills, memory, tools, files) via an OpenAI-compatible endpoint. Apps such as Codex, Cursor, Cline, Open WebUI and AnythingLLM can now call into a running Hermes agent—including one on a remote VPS—without leaving their UI. Recent update properly streams model reasoning so the client app can show thinking in real time.</p><p><b>Why it matters:</b> Removes the “leave the app and start over” friction. Builders can keep a single well-configured Hermes instance as the durable brain while using preferred front-ends.</p><p><a href="https://x.com/HermesWatcher/status/2102491889408811498">Post by @HermesWatcher</a></p>]]></description></item>
</channel></rss>
