<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Friday blog</title><description>Short, honest takes on AI coding agents and the harnesses that run them, from the team building Friday.</description><link>https://getfriday.dev/</link><item><title>Claude Code vs Codex vs OpenCode: which agent for which job?</title><link>https://getfriday.dev/blog/claude-code-vs-codex-vs-opencode/</link><guid isPermaLink="true">https://getfriday.dev/blog/claude-code-vs-codex-vs-opencode/</guid><description>We run all three coding agents daily. Claude Code wins on design and judgment, Codex on code generation and review, OpenCode on cost and model freedom. Here&apos;s the honest breakdown of when to use each.</description><pubDate>Mon, 06 Jul 2026 00:00:00 GMT</pubDate></item><item><title>What is an AI agent harness?</title><link>https://getfriday.dev/blog/what-is-an-ai-agent-harness/</link><guid isPermaLink="true">https://getfriday.dev/blog/what-is-an-ai-agent-harness/</guid><description>An AI agent harness runs coding agents like Claude Code and Codex toward a goal you set: it picks the agent, checks the work, and loops until it&apos;s done. Here&apos;s how that differs from a chatbot, a copilot, and a framework.</description><pubDate>Mon, 06 Jul 2026 00:00:00 GMT</pubDate></item></channel></rss>