<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Agents on Aleksandr Filippov</title><link>https://www.alexfeel.info/tags/agents/</link><description>Recent content in Agents on Aleksandr Filippov</description><generator>Hugo</generator><language>en</language><lastBuildDate>Thu, 16 Jul 2026 23:17:11 +0300</lastBuildDate><atom:link href="https://www.alexfeel.info/tags/agents/index.xml" rel="self" type="application/rss+xml"/><item><title>Dynamic Workflows: When the Agent Writes Its Own Harness</title><link>https://www.alexfeel.info/blog/post-6/</link><pubDate>Mon, 06 Jul 2026 05:00:00 +0000</pubDate><guid>https://www.alexfeel.info/blog/post-6/</guid><description>For a few weeks now, Claude Code has been writing its own multi-agent harness for each task &amp;ndash; and it quietly retired the orchestration code I used to maintain by hand. Here is what changed, the patterns worth knowing, and the research idea I think sits underneath it.</description></item><item><title>MCP Context Server: Persistent Memory for Your AI Coding Agents</title><link>https://www.alexfeel.info/blog/post-4/</link><pubDate>Sun, 19 Apr 2026 05:00:00 +0000</pubDate><guid>https://www.alexfeel.info/blog/post-4/</guid><description>Your AI coding agent loses its plan every time the context window compacts. MCP Context Server fixes that with persistent, searchable memory &amp;ndash; one Docker command and your agents remember everything.</description></item></channel></rss>