<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><title>SkillForge field notes</title><link>https://aidesignskills.dev/blog/</link><description>Practical lessons from building and testing reusable agent skills.</description><item><title>Building SkillForge with SiteSorted: two briefs, two outcomes</title><link>https://aidesignskills.dev/blog/building-skillforge-with-sitesorted/</link><guid>https://aidesignskills.dev/blog/building-skillforge-with-sitesorted/</guid><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><description>A first-hand account of two SiteSorted MCP builds: the original product page and the Vercel Docs-inspired documentation shell, including the manual work each required.</description></item><item><title>Portable agent skills in 2026: Claude, Codex, Gemini, Grok, v0, and Llama</title><link>https://aidesignskills.dev/blog/portable-agent-skills-2026/</link><guid>https://aidesignskills.dev/blog/portable-agent-skills-2026/</guid><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><description>A source-backed compatibility map for SKILL.md across frontier agent ecosystems, including native hosts, shared installers, and adapter-based runtimes.</description></item><item><title>How to test an agent skill: triggers, behavior, safety, and real outputs</title><link>https://aidesignskills.dev/blog/testing-agent-skills/</link><guid>https://aidesignskills.dev/blog/testing-agent-skills/</guid><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><description>A practical evaluation method for agent skills covering trigger precision, paired behavior, adversarial inputs, tool boundaries, and delivered artifacts.</description></item></channel></rss>
