<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/717995e2d75449998b639e4912f94d39&quot; frameborder=&quot;0&quot; width=&quot;1280&quot; height=&quot;960&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>960</height><width>1280</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>960</thumbnail_height><thumbnail_width>1280</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/717995e2d75449998b639e4912f94d39-1280047c85c9f666.gif</thumbnail_url><duration>216.363</duration><title>Building an AI Agent Integration Library</title><description>This Loom explains the author’s approach to an APM assignment by showing a step-by-step build plan for an AI integration library. They describe using an interactive HTML page to clarify and brainstorm the problem, then writing markdown files like principles, a project plan, and sprint checkpoints where decisions were made before code was added. The live site includes a library of seven applications with guides showing two key time estimates per app and walkthroughs for setup steps, paths, and sources. They also demonstrate a “Life Agent” feature that generates similar guides for new apps by checking official documentation and OAuth endpoints, and handling cases where OAuth does not exist, such as “Plausible.”</description></oembed>