<?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>Life Admin on Joey Gennari | Blog</title><link>http://gennari.industries/series/life-admin/</link><description>Recent content in Life Admin on Joey Gennari | Blog</description><generator>Hugo -- 0.147.3</generator><language>en</language><lastBuildDate>Tue, 28 Apr 2026 12:15:05 -0400</lastBuildDate><atom:link href="http://gennari.industries/series/life-admin/index.xml" rel="self" type="application/rss+xml"/><item><title>What Claude Remembers</title><link>http://gennari.industries/posts/what-claude-remembers/</link><pubDate>Tue, 28 Apr 2026 12:15:05 -0400</pubDate><guid>http://gennari.industries/posts/what-claude-remembers/</guid><description>&lt;p>The &lt;a href="http://gennari.industries/posts/my-life-runs-in-a-markdown-file/">first post&lt;/a> talked about the journal and check-in loop. The &lt;a href="http://gennari.industries/posts/skills-are-just-markdown/">second&lt;/a> talked about skills — markdown files Claude reads to know what to do. This last one is about how Claude remembers things across sessions.&lt;/p>
&lt;p>Claude Code ships with a feature called auto-memory: a small filesystem of facts the model can write to and read from, scoped to a project. When I correct Claude (&amp;ldquo;stop appending &amp;lsquo;Sent by Claude&amp;rsquo; to my texts&amp;rdquo;), that becomes a memory file. When I confirm a non-obvious decision (&amp;ldquo;yes, the bundled PR was the right call&amp;rdquo;), that&amp;rsquo;s also a memory. Anything I wouldn&amp;rsquo;t want to repeat to a fresh session — preferences, project context, pointers to where things live — goes there.&lt;/p></description></item><item><title>Skills Are Just Markdown</title><link>http://gennari.industries/posts/skills-are-just-markdown/</link><pubDate>Tue, 28 Apr 2026 11:17:47 -0400</pubDate><guid>http://gennari.industries/posts/skills-are-just-markdown/</guid><description>&lt;p>The &lt;a href="http://gennari.industries/posts/my-life-runs-in-a-markdown-file/">first post in this series&lt;/a> talked about &lt;code>/journal&lt;/code> and &lt;code>/checkin&lt;/code> — a daily journal and a 15-minute check-in loop that runs unattended on a schedule. I called them &amp;ldquo;skills&amp;rdquo; without explaining what that means. This post is about the mechanism.&lt;/p>
&lt;p>A skill is a markdown file. That&amp;rsquo;s the whole abstraction.&lt;/p>
&lt;p>Specifically: a directory under &lt;code>.claude/skills/&amp;lt;name&amp;gt;/&lt;/code> with a &lt;code>SKILL.md&lt;/code> inside. The directory name is the slash command. The contents of &lt;code>SKILL.md&lt;/code> are the instructions Claude reads when you invoke it. There&amp;rsquo;s no plugin API, no shell wrapper, no &amp;ldquo;register your command.&amp;rdquo; Drop a directory in, the command exists. Edit the file, the behavior changes.&lt;/p></description></item><item><title>My Life Runs in a Markdown File</title><link>http://gennari.industries/posts/my-life-runs-in-a-markdown-file/</link><pubDate>Mon, 27 Apr 2026 22:36:21 -0400</pubDate><guid>http://gennari.industries/posts/my-life-runs-in-a-markdown-file/</guid><description>&lt;p>I have a Claude Code-powered personal assistant that runs on a schedule. It reads my email, my calendar, my texts, the kids&amp;rsquo; school calendar, my plant care schedule, the weather, and my custody plan. It writes the day to a markdown file every morning and updates that file every 15 minutes throughout the work day. Most of the time it doesn&amp;rsquo;t tell me anything — that&amp;rsquo;s the point.&lt;/p>
&lt;p>The whole thing lives in a git repo I call &lt;code>life/&lt;/code>. About 6,000 lines of markdown, a handful of shell scripts, 17 slash commands. Here&amp;rsquo;s a partial list of what it does in a normal week, all without me asking:&lt;/p></description></item></channel></rss>