<?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/2c0e22669b004c049fe72de9252ecdac&quot; frameborder=&quot;0&quot; width=&quot;1670&quot; height=&quot;1252&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1252</height><width>1670</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1252</thumbnail_height><thumbnail_width>1670</thumbnail_width><thumbnail_url>https://cdn.loom.com/sessions/thumbnails/2c0e22669b004c049fe72de9252ecdac-f86ddb5cbf41e47e.gif</thumbnail_url><duration>120.704</duration><title>Ultra Project Manager Console AI Activity Tracking Demo</title><description>This Loom demos the Ultra Project Manager Console, an AI agent that can access team activity on a user computer at the OS level. It explains that the agent can read accessibility trees and perform OCR of the screen in real time, along with tracking which apps and projects team members are using. The live demo loads recent saved conversations, shows activity being logged, and uses Open Router to run a model while providing usage details. The console also uses MongoDB for its database, and the demo references topics like an Ultra PM console hackathon submission and a listed set of models or logs, including an incident where an API key was exposed.</description></oembed>