<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Agents on Chris Liatas</title><link>https://liatas.com/tags/ai-agents/</link><description>Recent content in AI Agents on Chris Liatas</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 21 Jul 2026 14:30:00 +0300</lastBuildDate><atom:link href="https://liatas.com/tags/ai-agents/index.xml" rel="self" type="application/rss+xml"/><item><title>Agent memory: Markdown, Git, discipline</title><link>https://liatas.com/posts/agent-memory-markdown-git/</link><pubDate>Tue, 21 Jul 2026 14:30:00 +0300</pubDate><guid>https://liatas.com/posts/agent-memory-markdown-git/</guid><description>&lt;div class="headerclaim"&gt;No prior experience with coding agents is assumed beyond having tried one (Claude Code, Codex, Cursor and friends); Git basics help. The setup shown is one I actually run daily; examples are sanitized — structure and scripts are real, note contents are illustrative.&lt;/div&gt;

&lt;p&gt;Coding agents are usually sharp inside a session and forgetful between them. Every new thread starts from zero — the pitfall hit last Tuesday, the reason a dependency got pinned, the deployment decision everyone agreed on — gone, unless someone repeats it again. The popular fix is refreshingly low-tech: write things down in Markdown files, and let the agent read them back.&lt;/p&gt;</description></item></channel></rss>