<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>agent on The Offcuts | Halle Winkler</title><link>https://theoffcuts.org/tags/agent/</link><description>Recent content in agent on The Offcuts | Halle Winkler</description><generator>Hugo -- gohugo.io</generator><language>en-us</language><lastBuildDate>Thu, 01 Oct 2026 12:00:00 +0100</lastBuildDate><atom:link href="https://theoffcuts.org/tags/agent/index.xml" rel="self" type="application/rss+xml"/><item><title>Local LLM vulnerability detection and sandboxing on underprovisioned Silicon Part 2 of 3: The Sandboxed Local Agent</title><link>https://theoffcuts.org/posts/local-llm-vulnerability-detection-and-sandboxing-on-underprovisioned-silicon-part-2/</link><pubDate>Thu, 01 Oct 2026 12:00:00 +0100</pubDate><guid>https://theoffcuts.org/posts/local-llm-vulnerability-detection-and-sandboxing-on-underprovisioned-silicon-part-2/</guid><description>Local LLM vulnerability detection and sandboxing on underprovisioned Silicon, a 3 part series. # Part 2 of 3: The Sandboxed Local Agent # October 1st, 2026 To the code →
In part 1 we got our local oMLX server running and had a chat with it.
In part 2, we&amp;rsquo;re going to set up a local open-source agent which can talk to this server for its LLM, and we&amp;rsquo;re going to sandbox it to the point that we are satisfied that we don&amp;rsquo;t have to rely on the agentic sandbox and don&amp;rsquo;t have to read into the motivations of its developers (which might be its agents?</description></item></channel></rss>